Examples gallery¶

These are runnable examples from binoc's test suite. Each example links to its source folder on GitHub, tells you whether it needs any extra setup, gives you the exact command to run, and shows the Markdown changelog binoc is expected to print.

Binoc currently ships 82 shared examples in this gallery.

One-time setup¶

Clone the repository and materialize the archive-based fixtures once:

git clone https://github.com/harvard-lil/binoc
cd binoc
just materialize

At a glance¶

Example	What it shows	Example output	Setup
`binary-byte-range-localized`	An opaque binary changes in one localized byte region; CDC byte-range localization reports the changed range and unchan…	payload.bin: 1 changed byte range; 66.667% unchanged; first range left [65,536, 131,072) to right […	Default pipeline
`binary-fallback-diagnostic`	Unknown file type compared by the binary fallback emits a suggestion	data.parquet: Binary content changed; 1 extracted string added, 1 extracted string removed	Default pipeline
`binary-strings-fallback`	Two opaque binary blobs with differing hashes. The change is hash-driven (binoc.content-changed), and an additive extra…	firmware.bin: Binary content changed; 2 extracted strings added, 2 extracted strings removed	Default pipeline
`content-sniff-extensionless-tz`	Extensionless IANA tz source files are sniffed as text and disclose that inference; a binary sibling in the same bundle…	asia: 1 line added	Default pipeline
`csv-auto-detected-pipe-dialect`	Extensionless tabular input auto-detects a pipe dialect, discloses the inference, and still reports keyed cell changes	records: 1 row modified by key	Custom config
`csv-auto-key-resort`	Auto-detected CSV row keys turn a high positional-churn re-sort into a keyed cell change	data.csv: 1 row modified by key	Default pipeline
`csv-cell-changes`	Individual cell values changed	data.csv: 2 cells changed	Default pipeline
`csv-column-addition`	New column added	data.csv: Column added: 'email'	Default pipeline
`csv-column-removal`	Column removed	data.csv: Column removed: 'city'	Default pipeline
`csv-column-rename-near-miss`	A weak content overlap should remain a column add plus remove	data.csv: Column added: 'status'; Column removed: 'legacy'	Default pipeline
`csv-column-rename-reorder`	A renamed CSV column also moves position	data.csv: Column renamed: 'status' -> 'state'; Columns reordered	Default pipeline
`csv-column-reorder`	Columns shuffled, content identical	data.csv: Columns reordered	Default pipeline
`csv-declared-pipe-dialect`	Per-path declared pipe dialect parses a .txt table silently and reports keyed cell changes	data.txt: 1 row modified by key	Custom config
`csv-disjoint-cohort-guardrail`	Disjoint CSV cohorts trip the high-churn guardrail instead of publishing a fictional per-cell diff	data.csv: row correspondence uncertain; 1 changed-cell fraction	Default pipeline
`csv-distribution-shift`	Numeric column distribution shifts with keyed row matching	data.csv: 4 rows modified by key	Custom config
`csv-keyed-null-duplicate`	Configured CSV row keys surface null and duplicate key diagnostics	data.csv: 14 cells changed	Custom config
`csv-keyed-row-diff`	Configured CSV row keys match reordered rows and report keyed row/cell changes	data.csv: 1 row added; 1 row removed; 1 row modified by key	Custom config
`csv-mid-row-insertion`	A mid-table row insertion compacts while column reorder/addition rules remain independent	data.csv: Column added: 'email'; Columns reordered; 1 row added	Default pipeline
`csv-mixed-changes`	Multiple change types	data.csv: Column added: 'email'; Columns reordered; 1 row added	Default pipeline
`csv-numeric-rounding`	CSV numeric cells rounded to a common modulus and representation-only numeric differences ignored	data.csv: Rounded 3 cells in 'population' to nearest 1000	Default pipeline
`csv-rename-modify`	CSV renamed and modified: detected as a single move by fuzzy correlation	data_v2.csv:	Default pipeline
`csv-row-addition`	New rows appended	data.csv: 2 rows added	Default pipeline
`csv-row-insertion-suppression-sentinels`	Inserted rows with disclosure sentinels in count cells are reported as row additions, not value suppression	data.csv: 2 rows added	Default pipeline
`csv-row-removal`	Rows removed from CSV	data.csv: 2 rows removed	Default pipeline
`csv-sorted-row-fallback`	CSV rows with no unique key use sorted row-content alignment so a pure re-sort is not reported as cell churn	data.csv: rows reordered; cell values unchanged under inferred row alignment	Default pipeline
`csv-stacked-tables`	Detects two logical tables stacked in one messy CSV	data.csv/>table_2: 1 row added	Default pipeline
`csv-to-tsv-reformat`	Table reformatted from CSV to TSV with row edits: detected as one reformatted-and-modified table, not remove + add	data.tsv:	Default pipeline
`csv-value-suppression`	CSV cells replaced with disclosure suppression sentinels	data.csv: Suppressed 3 cells in 'count'	Default pipeline
`csv-value-suppression-custom-sentinel`	CSV cells replaced with a dataset-configured disclosure suppression sentinel	data.csv: 1 cell changed; Suppressed 2 cells in 'count'	Custom config
`csv-verbosity-full`	Markdown full verbosity renders every captured changed-cell example.	data.csv: 5 cells changed	Custom config
`csv-vintage-benchmark`	A 'vintage' reader compares two editions of the same published dataset and wants the structural story (a column appeared, a category vocabulary shifted) surfaced above the bulk data churn they intend to ignore.	facilities.csv: Column added: 'region'; 1 cell changed	Custom config
`directory-file-copy`	New file with same content as an existing unchanged file detected as a copy	duplicate.txt: Copied from original.txt	Default pipeline
`directory-nested`	Subdirectories with mixed changes	data/records.csv: 1 row added	Default pipeline
`directory-nested-with-tar`	Shows binoc diffing a tar archive and a plain directory that contain overlapping internal paths.	data.tar.gz/>records.csv: 1 cell changed	Default pipeline
`enforcement-actions-merge-years`	Per-year CSVs merged row-wise into one file; detected as a clean partition merge (CFM-72)	actions_2023.csv, actions_2024.csv merged into actions.csv	Default pipeline
`file-correspondence-container`	Config declares a correspondence between renamed zip containers	archive.zip: Moved from data.zip	Custom config
`file-correspondence-scheme`	Config declares that a state CSV moved into a new directory scheme is the same logical file	by-state: Added	Custom config
`file-correspondence-token`	Config declares that year-stamped CSV filenames are the same logical file	running_list_as_of_2023.csv:	Custom config
`folder-move-nested`	Detects a whole-folder rename and rolls many file moves up into one folder-move entry.	documentation: Moved from docs	Default pipeline
`folder-move-partial`	Detects a mostly-moved folder rename and preserves only the added/removed/modified remainder entries beneath it.	FoodData_Central_csv_2026-04-30: Added	Default pipeline
`geojson-feature-cell-change`	A GeoJSON FeatureCollection where one feature's property changes; transcoded to a tabular artifact with the geometry as…	places.geojson: 1 cell changed	Default pipeline
`gzip-inner-dispatch`	Gzipped CSV and text are decompressed and redispatched under their inner names	census.txt.gz/>census.txt: 1 line added; 1 line removed	Default pipeline
`ini-value-change`	An INI value changes; transcoded to a structured_document and reported as a value change	config.ini: $.replicas: "3" -> "5"	Default pipeline
`json-array-order-significant`	JSON array order changes are semantic content changes in stage 1	metadata.json: $.ids[1]: 2 -> 3; $.ids[2]: 3 -> 2	Default pipeline
`json-key-order-reexport`	JSON object key order and pretty-printing changed without semantic value changes	metadata.json: Document serialization changed	Default pipeline
`json-keyed-row-diff`	Configured JSON record keys match reordered rows and report keyed row/cell changes	data.json: 1 row added; 1 row removed; 1 row modified by key	Custom config
`json-records-cell-change`	JSON array of like-shaped objects parsed as a typed table; numeric cell values change	data.json: 2 cells changed	Default pipeline
`json-records-nested-value`	JSON records with a nested object cell; the nested value changes and is reported as a single equality-based cell edit (…	people.json: 1 cell changed	Default pipeline
`json-records-path-stix-objects`	STIX-shaped JSON bundle records under $.objects are parsed as a keyed table	enterprise.stix.json: 1 row modified by key	Custom config
`json-records-type-only-column`	Typed JSON records re-serialized with one numeric column as strings; canonical values match	data.json: Column type changed: 'year' number -> string	Default pipeline
`jsonl-row-addition`	JSONL stream of like-shaped objects parsed as a table; a record is appended	events.jsonl: 1 row added	Default pipeline
`jsonld-value-change`	A .jsonld file with no declared media type parses as a structured document tagged format=jsonld; a value change is repo…	person.jsonld: $.jobTitle: "Mathematician" -> "Computer Scientist"	Default pipeline
`kitchen-sink`	Runs text, CSV, archive, move, and copy detection together in one end-to-end example.	archive.tar.gz/>inventory.csv: 1 row added	Default pipeline
`nasa-gistemp-header-line`	NASA GISTEMP-style table with a title line before the real header	GLB.Ts+dSST.csv: 1 row modified by key	Custom config
`observations-repartition-equal-arity`	Equal-arity N→M repartition: 2 tables grouped by region become 2 tables grouped by year, every row preserved exactly bu…	observations_2024.csv:	Default pipeline
`observations-split-by-year`	One CSV split row-wise into per-year files; detected as a clean partition split (CFM-72)	observations.csv split into observations_2024.csv, observations_2025.csv	Default pipeline
`observations-split-residual`	A would-be split missing one row: partition declines (not complete), emits binoc.possible_split, and degrades to honest…	observations_2024.csv:	Default pipeline
`ofac-sdn-headerless-position-key`	OFAC SDN-style 12-column headerless CSV keyed by ent_num in column 1	SDN.CSV: 1 row modified by key	Custom config
`per-path-dispatch-override`	Per-path dispatch overrides promote extensionless CSV content before tabular row keying	forced: 1 row modified by key	Custom config
`single-file-add`	File present in B but not A	new_file.txt: Added	Default pipeline
`single-file-modify-binary`	Binary file, different hash	data.bin: 1 edit	Default pipeline
`single-file-modify-csv`	CSV file compared directly (file-to-file, not via directory)	data.csv: 1 row added	Default pipeline
`single-file-modify-text`	Text file with line-level changes	story.txt: 2 lines added; 1 line removed	Default pipeline
`single-file-modify-text-root`	Text file compared directly (file-to-file, not via directory)	story.txt: 2 lines added; 1 line removed	Default pipeline
`single-file-remove`	File present in A but not B	removed_file.txt: Removed	Default pipeline
`stacked-csv-broken-out`	Stacked-CSV tables broken out into one file per table; whole-table rehoming (reshape + 1:1), NOT a partition split (CFM…	changes.csv: Moved from report.csv/>table_1	Default pipeline
`tar-nested`	Nested tar.gz containing CSV	outer.tar.gz/>inner.tar.gz/>data.csv: 1 row added	Default pipeline
`tar-simple`	Tar.gz archive with changes inside	archive.tar.gz/>data.csv: 1 row added	Default pipeline
`text-rename-modify`	Text file renamed and modified: detected as a single move by fuzzy correlation	meeting-notes-v2.txt:	Default pipeline
`toml-value-change`	A TOML value changes; transcoded to a structured_document and reported as a value change	config.toml: $.replicas: 3 -> 5	Default pipeline
`tree-wide-correlation`	Shows tree-wide move and copy detection across nested zip boundaries, including one-to-many copies and many-to-one moves.	gamma-renamed.txt: Moved from outer.zip/>inner.zip/>gamma.txt	Default pipeline
`trivial-identical`	Two identical directories → empty changeset	# Changelog: snapshot-a → snapshot-b	Default pipeline
`trivial-identical-csv`	Two identical CSV files → no changes reported	# Changelog: snapshot-a → snapshot-b	Default pipeline
`tsv-cell-changes`	Tab-delimited file parses into real columns and reports cell changes	data.tsv: 2 cells changed	Default pipeline
`yaml-value-change`	A YAML scalar value changes; transcoded to a structured_document and reported as a value change	config.yaml: $.replicas: 3 -> 5	Default pipeline
`zip-declared-container`	Config declares a correspondence between nested zip containers and preserves inner CSV content detail	outer.zip/>records.zip: Moved from outer.zip/>records-old.zip	Custom config
`zip-json-key-order-reexport`	JSON files inside zip expansion get parsed and rendered as serialization-only changes	archive.zip/>metadata.json: Document serialization changed	Default pipeline
`zip-nested`	Nested zip containing CSV	outer.zip/>inner.zip/>data.csv: 1 row added	Default pipeline
`zip-rename-contents-rewritten`	Documents a known gap — a renamed zip whose children were all renamed AND rewritten (no content similarity) yields unpa…	data.zip: Removed	Default pipeline
`zip-rename-identical`	Zip archive renamed with identical contents; bottom-up roll-up of the inner clean file moves compacts the pair into a s…	archive.zip: Moved from data.zip	Default pipeline
`zip-rename-inner-rename-edit`	Zip archive renamed while its only child was renamed and had one cell edited; the modified move counts as roll-up evide…	archive.zip: Moved from data.zip	Default pipeline
`zip-simple`	Zipped files with changes inside	archive.zip/>data.txt: 1 line added; 1 line removed	Default pipeline

binary-byte-range-localized¶

An opaque binary changes in one localized byte region; CDC byte-range localization reports the changed range and unchan…

Browse source: binary-byte-range-localized
Tags: modify, binary, byte-range
Snapshots: snapshot-a has 1 file — payload.bin; snapshot-b has 1 file — payload.bin

Run it:

binoc diff \
  ./test-vectors-materialized/binary-byte-range-localized/snapshot-a \
  ./test-vectors-materialized/binary-byte-range-localized/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **payload.bin**: 1 changed byte range; 66.667% unchanged; first range left [65,536, 131,072) to right [65,536, 131,072)

binary-fallback-diagnostic¶

Unknown file type compared by the binary fallback emits a suggestion

Browse source: binary-fallback-diagnostic
Tags: modify, binary, diagnostics
Snapshots: snapshot-a has 1 file — data.parquet; snapshot-b has 1 file — data.parquet

Run it:

binoc diff \
  ./test-vectors-materialized/binary-fallback-diagnostic/snapshot-a \
  ./test-vectors-materialized/binary-fallback-diagnostic/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.parquet**: Binary content changed; 1 extracted string added, 1 extracted string removed
  - Extracted strings added
    - 'after!\n'
  - Extracted strings removed
    - 'before\n'

binary-strings-fallback¶

Two opaque binary blobs with differing hashes. The change is hash-driven (binoc.content-changed), and an additive extra…

Browse source: binary-strings-fallback
Tags: modify, binary, strings
Snapshots: snapshot-a has 1 file — firmware.bin; snapshot-b has 1 file — firmware.bin

Run it:

binoc diff \
  ./test-vectors-materialized/binary-strings-fallback/snapshot-a \
  ./test-vectors-materialized/binary-strings-fallback/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **firmware.bin**: Binary content changed; 2 extracted strings added, 2 extracted strings removed
  - Extracted strings added
    - 'build-beta'
    - 'version=2.0.0'
  - Extracted strings removed
    - 'build-alpha'
    - 'version=1.0.0'

content-sniff-extensionless-tz¶

Extensionless IANA tz source files are sniffed as text and disclose that inference; a binary sibling in the same bundle…

Browse source: content-sniff-extensionless-tz
Tags: text, binary, inference, content-sniff
Snapshots: snapshot-a has 4 files — asia, europe, northamerica, tzdata.binless; snapshot-b has 4 files — asia, europe, northamerica, tzdata.binless

Run it:

binoc diff \
  ./test-vectors-materialized/content-sniff-extensionless-tz/snapshot-a \
  ./test-vectors-materialized/content-sniff-extensionless-tz/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **asia**: 1 line added
  - Content type inference: treated asia as text (content sniff, no extension)
  - Line changes
    - line 2: 'Zone    Asia/Tokyo  9:18:59 -   LMT 1887 Dec 31 15:00u' -> 'Rule    Japan   1949    only    -   Apr Sat>=1  24:00   1:00    D'
- **europe**: 1 line added
  - Content type inference: treated europe as text (content sniff, no extension)
  - Line changes
    - line 2: 'Zone    Europe/Paris    0:09:21 -   LMT 1891 Mar 16' -> 'Rule    EU  1996    max -   Oct lastSun   1:00u   0   -'
- **northamerica**: 1 line added
  - Content type inference: treated northamerica as text (content sniff, no extension)
  - Line changes
    - line 2: 'Zone    America/New_York -4:56:02 -   LMT 1883 Nov 18 17:00u' -> 'Rule    US  2007    max -   Mar Sun>=8    2:00    1:00    D'
- **tzdata.binless**: 1 edit

csv-auto-detected-pipe-dialect¶

Extensionless tabular input auto-detects a pipe dialect, discloses the inference, and still reports keyed cell changes

Browse source: csv-auto-detected-pipe-dialect
Tags: csv, dialect, pipe, inference, extensionless
Snapshots: snapshot-a has 1 file — records; snapshot-b has 1 file — records
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-auto-detected-pipe-dialect.yaml:

dataset:
  paths:
    - match: records
      content_type: text/csv
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/csv-auto-detected-pipe-dialect/snapshot-a \
  ./test-vectors-materialized/csv-auto-detected-pipe-dialect/snapshot-b \
  --config /tmp/csv-auto-detected-pipe-dialect.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **records**: 1 row modified by key
  - Dialect provenance: detected `|`-delimited, no quoting, newline LF
  - Changed cells
    - key id '1', column 'value': 'old' -> 'new'

csv-auto-key-resort¶

Auto-detected CSV row keys turn a high positional-churn re-sort into a keyed cell change

Browse source: csv-auto-key-resort
Tags: csv, auto-key, row-reorder, cell-change
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-auto-key-resort/snapshot-a \
  ./test-vectors-materialized/csv-auto-key-resort/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 1 row modified by key
  - Changed cells
    - key id 'brfss-003', column 'rate': '14.8' -> '15.0'

## Suggestions

- inferred row identity column 'id' from unique values with 100% overlap (`binoc.write.tabular`) [binoc.tabular_auto_key]

csv-cell-changes¶

Individual cell values changed

Browse source: csv-cell-changes
Tags: csv, cell-change
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-cell-changes/snapshot-a \
  ./test-vectors-materialized/csv-cell-changes/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 2 cells changed
  - Changed cells
    - row 1, column 'score': '85' -> '92'
    - row 2, column 'score': '90' -> '88'

csv-column-addition¶

New column added

Browse source: csv-column-addition
Tags: csv, column-addition, schema
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-column-addition/snapshot-a \
  ./test-vectors-materialized/csv-column-addition/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column added: 'email'
  - Set Headers: from: ["name","age"]; to: ["name","age","email"]
  - Add Column: name: 'email'; values: {"total_values":2,"truncated":false,"values":["alice@test.com","bob@test.com"]}

csv-column-removal¶

Column removed

Browse source: csv-column-removal
Tags: csv, column-removal, schema
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-column-removal/snapshot-a \
  ./test-vectors-materialized/csv-column-removal/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column removed: 'city'
  - Set Headers: from: ["name","age","city"]; to: ["name","age"]
  - Remove Column: name: 'city'; values: {"total_values":2,"truncated":false,"values":["NYC","LA"]}

csv-column-rename-near-miss¶

A weak content overlap should remain a column add plus remove

Browse source: csv-column-rename-near-miss
Tags: csv, column-addition, column-removal
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-column-rename-near-miss/snapshot-a \
  ./test-vectors-materialized/csv-column-rename-near-miss/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column added: 'status'; Column removed: 'legacy'
  - Set Headers: from: ["id","legacy"]; to: ["id","status"]
  - Add Column: name: 'status'; values: {"total_values":4,"truncated":false,"values":["alpha","draft","hold","closed"]}
  - Remove Column: name: 'legacy'; values: {"total_values":4,"truncated":false,"values":["alpha","beta","gamma","delta"]}

csv-column-rename-reorder¶

A renamed CSV column also moves position

Browse source: csv-column-rename-reorder
Tags: csv, column-rename, column-reorder
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-column-rename-reorder/snapshot-a \
  ./test-vectors-materialized/csv-column-rename-reorder/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column renamed: 'status' -> 'state'; Columns reordered

csv-column-reorder¶

Columns shuffled, content identical

Browse source: csv-column-reorder
Tags: csv, column-reorder, clerical
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-column-reorder/snapshot-a \
  ./test-vectors-materialized/csv-column-reorder/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Columns reordered

csv-declared-pipe-dialect¶

Per-path declared pipe dialect parses a .txt table silently and reports keyed cell changes

Browse source: csv-declared-pipe-dialect
Tags: csv, dialect, pipe, per-path
Snapshots: snapshot-a has 1 file — data.txt; snapshot-b has 1 file — data.txt
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-declared-pipe-dialect.yaml:

dataset:
  paths:
    - match: data.txt
      content_type: text/csv
      row_identity:
        columns:
          - id
      dialect:
        delimiter: |

Run it:

binoc diff \
  ./test-vectors-materialized/csv-declared-pipe-dialect/snapshot-a \
  ./test-vectors-materialized/csv-declared-pipe-dialect/snapshot-b \
  --config /tmp/csv-declared-pipe-dialect.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.txt**: 1 row modified by key
  - Changed cells
    - key id '1', column 'value': 'old' -> 'new'

csv-disjoint-cohort-guardrail¶

Disjoint CSV cohorts trip the high-churn guardrail instead of publishing a fictional per-cell diff

Browse source: csv-disjoint-cohort-guardrail
Tags: csv, guardrail, high-churn
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-disjoint-cohort-guardrail/snapshot-a \
  ./test-vectors-materialized/csv-disjoint-cohort-guardrail/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: row correspondence uncertain; 1 changed-cell fraction

## Suggestions

- these two tables don't appear to correspond row-for-row (100% changed cells; candidate key columns: id, rate, county) (`data.csv`) [binoc.tabular_high_churn]
  - use `binoc extract CHANGESET "data.csv" content`

csv-distribution-shift¶

Numeric column distribution shifts with keyed row matching

Browse source: csv-distribution-shift
Tags: csv, statistics, row-identity
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-distribution-shift.yaml:

dataset:
  defaults:
    row_identity:
      columns:
        - id

Run it:

binoc diff \
  ./test-vectors-materialized/csv-distribution-shift/snapshot-a \
  ./test-vectors-materialized/csv-distribution-shift/snapshot-b \
  --config /tmp/csv-distribution-shift.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 4 rows modified by key
  - Changed cells (showing 3 of 5)
    - changed cells by column: score 4, label 1
    - key id '1', column 'score': '10' -> '12'
    - key id '2', column 'label': 'beta' -> 'beta2'
    - key id '2', column 'score': '20' -> '35'

csv-keyed-null-duplicate¶

Configured CSV row keys surface null and duplicate key diagnostics

Browse source: csv-keyed-null-duplicate
Tags: csv, keyed, null-key, duplicate-key
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-keyed-null-duplicate.yaml:

dataset:
  defaults:
    row_identity:
      on_null_key: diagnostic
      on_duplicate_key: diagnostic
  paths:
    - match: data.csv
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/csv-keyed-null-duplicate/snapshot-a \
  ./test-vectors-materialized/csv-keyed-null-duplicate/snapshot-b \
  --config /tmp/csv-keyed-null-duplicate.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 14 cells changed
  - Changed cells (showing 3 of 14)
    - changed cells by column: id 4, name 5, score 5
    - row 1, column 'id': 'a' -> 'b'
    - row 1, column 'name': 'Alice' -> 'Bob'
    - row 1, column 'score': '10' -> '21'

## Warnings

- configured row keys had null values; fell back to positional row comparison (`data.csv`) [binoc.keyed_row_identity_degraded]
  - use `binoc extract CHANGESET "data.csv" content`

csv-keyed-row-diff¶

Configured CSV row keys match reordered rows and report keyed row/cell changes

Browse source: csv-keyed-row-diff
Tags: csv, keyed, row-addition, row-removal, cell-change
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-keyed-row-diff.yaml:

dataset:
  paths:
    - match: data.csv
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/csv-keyed-row-diff/snapshot-a \
  ./test-vectors-materialized/csv-keyed-row-diff/snapshot-b \
  --config /tmp/csv-keyed-row-diff.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 1 row added; 1 row removed; 1 row modified by key
  - Changed cells
    - key id 'p2', column 'price': '20' -> '25'
  - Rows added
    - key id 'p4': 'p4', 'Delta', '40'
  - Rows removed
    - key id 'p3': 'p3', 'Gamma', '30'

csv-mid-row-insertion¶

A mid-table row insertion compacts while column reorder/addition rules remain independent

Browse source: csv-mid-row-insertion
Tags: csv, row-addition, column-reorder, column-addition, lcs, compaction
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-mid-row-insertion/snapshot-a \
  ./test-vectors-materialized/csv-mid-row-insertion/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column added: 'email'; Columns reordered; 1 row added
  - Rows added
    - row 2: 'LA', 'Bob', '25'
  - Add Column: name: 'email'; values: {"total_values":3,"truncated":false,"values":["alice@example.test","bob@example.test","charlie@example.test"]}

csv-mixed-changes¶

Multiple change types

Browse source: csv-mixed-changes
Tags: csv, column-reorder, column-addition, row-addition
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-mixed-changes/snapshot-a \
  ./test-vectors-materialized/csv-mixed-changes/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Column added: 'email'; Columns reordered; 1 row added
  - Rows added
    - row 3: 'SF', 'Charlie', '35'
  - Add Column: name: 'email'; values: {"total_values":3,"truncated":false,"values":["a@test.com","b@test.com","c@test.com"]}

csv-numeric-rounding¶

CSV numeric cells rounded to a common modulus and representation-only numeric differences ignored

Browse source: csv-numeric-rounding
Tags: csv, cell-change, value-rounding
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-numeric-rounding/snapshot-a \
  ./test-vectors-materialized/csv-numeric-rounding/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Rounded 3 cells in 'population' to nearest 1000

csv-rename-modify¶

CSV renamed and modified: detected as a single move by fuzzy correlation

Browse source: csv-rename-modify
Tags: csv, fuzzy-move, rename-modify
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data_v2.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-rename-modify/snapshot-a \
  ./test-vectors-materialized/csv-rename-modify/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data_v2.csv**:
  - Moved from data.csv
  - Column added: 'email'
  - Set Headers: from: ["name","age","city"]; to: ["name","age","city","email"]
  - Add Column: name: 'email'; values: {"total_values":3,"truncated":false,"values":["alice@test.com","bob@test.com","carol@test.com"]}

csv-row-addition¶

New rows appended

Browse source: csv-row-addition
Tags: csv, row-addition
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-row-addition/snapshot-a \
  ./test-vectors-materialized/csv-row-addition/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 2 rows added
  - Rows added
    - row 2: 'Bob', '25'
    - row 3: 'Charlie', '35'

csv-row-insertion-suppression-sentinels¶

Inserted rows with disclosure sentinels in count cells are reported as row additions, not value suppression

Browse source: csv-row-insertion-suppression-sentinels
Tags: csv, row-addition, value-suppression-regression
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-row-insertion-suppression-sentinels/snapshot-a \
  ./test-vectors-materialized/csv-row-insertion-suppression-sentinels/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 2 rows added
  - Rows added
    - row 2: 'Beta', '*'
    - row 3: 'Gamma', '(D)'

csv-row-removal¶

Rows removed from CSV

Browse source: csv-row-removal
Tags: csv, row-removal
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-row-removal/snapshot-a \
  ./test-vectors-materialized/csv-row-removal/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 2 rows removed
  - Rows removed
    - row 2: 'Bob', '25'
    - row 3: 'Charlie', '35'

csv-sorted-row-fallback¶

CSV rows with no unique key use sorted row-content alignment so a pure re-sort is not reported as cell churn

Browse source: csv-sorted-row-fallback
Tags: csv, row-reorder, sorted-fallback
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-sorted-row-fallback/snapshot-a \
  ./test-vectors-materialized/csv-sorted-row-fallback/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: rows reordered; cell values unchanged under inferred row alignment

csv-stacked-tables¶

Detects two logical tables stacked in one messy CSV

Browse source: csv-stacked-tables
Tags: csv, stacked-tables, row-addition
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-stacked-tables/snapshot-a \
  ./test-vectors-materialized/csv-stacked-tables/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv/>table_2**: 1 row added
  - Rows added
    - row 12: '761012', 'Mu', 'Mu Pharma'

csv-to-tsv-reformat¶

Table reformatted from CSV to TSV with row edits: detected as one reformatted-and-modified table, not remove + add

Browse source: csv-to-tsv-reformat
Tags: csv, tsv, reformat, serialization-change, tabular-pair
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.tsv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-to-tsv-reformat/snapshot-a \
  ./test-vectors-materialized/csv-to-tsv-reformat/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.tsv**:
  - Moved from data.csv
  - 1 row added; 1 cell changed
  - Dialect provenance: detected tab-delimited, no quoting, newline LF
  - Changed cells
    - row 2, column 'age': '25' -> '26'
  - Rows added
    - row 4: 'Dave', '41', 'Austin'

csv-value-suppression¶

CSV cells replaced with disclosure suppression sentinels

Browse source: csv-value-suppression
Tags: csv, cell-change, value-suppression
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/csv-value-suppression/snapshot-a \
  ./test-vectors-materialized/csv-value-suppression/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: Suppressed 3 cells in 'count'

csv-value-suppression-custom-sentinel¶

CSV cells replaced with a dataset-configured disclosure suppression sentinel

Browse source: csv-value-suppression-custom-sentinel
Tags: csv, cell-change, value-suppression, dataset-config
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/csv-value-suppression-custom-sentinel.yaml:

dataset:
  reduced_precision:
    suppression_sentinels:
      - N/A
      - ""

Run it:

binoc diff \
  ./test-vectors-materialized/csv-value-suppression-custom-sentinel/snapshot-a \
  ./test-vectors-materialized/csv-value-suppression-custom-sentinel/snapshot-b \
  --config /tmp/csv-value-suppression-custom-sentinel.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 1 cell changed; Suppressed 2 cells in 'count'
  - Changed cells
    - row 3, column 'rate': '8.9' -> '9.1'

csv-verbosity-full¶

Markdown full verbosity renders every captured changed-cell example.

Browse source: csv-verbosity-full
Tags: csv, cell-change, verbosity
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv
Setup: This example sets output.markdown.verbosity: full so the changelog prints every captured changed-cell example instead of the default capped sample. Save this dataset config as /tmp/csv-verbosity-full.yaml:

output:
  markdown:
    verbosity: full

Run it:

binoc diff \
  ./test-vectors-materialized/csv-verbosity-full/snapshot-a \
  ./test-vectors-materialized/csv-verbosity-full/snapshot-b \
  --config /tmp/csv-verbosity-full.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 5 cells changed
  - Sources
    - data.csv (from, modify, binoc.pair.name)
  - Changed cells
    - row 1, column 'score': '10' -> '11'
    - row 2, column 'score': '20' -> '21'
    - row 3, column 'score': '30' -> '31'
    - row 4, column 'score': '40' -> '41'
    - row 5, column 'score': '50' -> '51'

csv-vintage-benchmark¶

A 'vintage' reader compares two editions of the same published dataset and wants the structural story (a column appeared, a category vocabulary shifted) surfaced above the bulk data churn they intend to ignore.

Browse source: csv-vintage-benchmark
Tags: csv, vintage, metadata, benchmark
Snapshots: snapshot-a has 2 files — facilities.csv, inspections.csv; snapshot-b has 2 files — facilities.csv, inspections.csv
Setup: The dataset is a yearly facilities register published as a small directory of CSVs. Between the two editions:
facilities.csv gains a region column (schema change) and one row's status moves to a brand-new category value, decommissioned (a vocabulary shift — the set of distinct values in a categorical column grew).
inspections.csv changes only in its data: several scores are edited and two rows are appended. This is exactly the churn a vintage reader does not want to read.

The markdown config models the vintage stance as significance: schema/structural tags are the high-priority group, bulk cell/row tags the low-priority group. Because classify_tags promotes a node to the highest-priority group among its tags, facilities.csv (which carries both schema and cell tags) floats up to "Schema & vocabulary changes" while the pure-data inspections.csv sinks to "Bulk data updates". That file-granularity separation is the best vintage view binoc offers today.

WHAT THIS BENCHMARK IS FOR — the gap between today's output (see expected-output/changelog.snap) and the target (see VINTAGE-IDEAL.md):

Within-node significance. facilities.csv's region addition and its status cell edit live on one node, so they cannot be separated: the vintage reader still sees the cell bullet. There is no config-driven edit-level drop/keep (only EditProjection.visible, set by writers).
Vocabulary as a first-class change. The active -> decommissioned shift is reported as an ordinary binoc.cell-change, not as "the status vocabulary gained a value". Columns are not first-class nodes and distinct-value-set diffing does not exist.
Summary statistics. inspections.csv is rendered as full cell/row detail, not as a one-line vintage statistic ("142 -> 144 rows, 3 cells changed"). The Summary/GlobalClaim seams exist to carry such a fact; no rule emits one.

This vector is a kept benchmark, not a feature. It is expected to PASS against current output; as the vintage story improves, update the snapshot and watch it converge on VINTAGE-IDEAL.md. See docs/adr for the design rationale. Save this dataset config as /tmp/csv-vintage-benchmark.yaml:

output:
  markdown:
    groups:
      - heading: Schema & vocabulary changes
        tags:
          - binoc.schema-change
          - binoc.column-addition
          - binoc.column-removal
          - binoc.column-rename
          - binoc.metadata.value-label-set
      - heading: Bulk data updates
        tags:
          - binoc.cell-change
          - binoc.row-addition
          - binoc.row-removal

Run it:

binoc diff \
  ./test-vectors-materialized/csv-vintage-benchmark/snapshot-a \
  ./test-vectors-materialized/csv-vintage-benchmark/snapshot-b \
  --config /tmp/csv-vintage-benchmark.yaml

Result:

# Changelog: snapshot-a → snapshot-b

## Schema & vocabulary changes

- **facilities.csv**: Column added: 'region'; 1 cell changed
  - Changed cells
    - row 2, column 'status': 'active' -> 'decommissioned'
  - Set Headers: from: ["facility_id","name","status"]; to: ["facility_id","name","status","region"]
  - Add Column: name: 'region'; values: {"total_values":4,"truncated":false,"values":["north","east","west","south"]}

## Bulk data updates

- **inspections.csv**: 2 rows added; 3 cells changed
  - Changed cells
    - row 1, column 'score': '82' -> '85'
    - row 3, column 'score': '90' -> '91'
    - row 4, column 'score': '68' -> '70'
  - Rows added
    - row 5: 'I104', 'F001', '88'
    - row 6: 'I105', 'F002', '73'

directory-file-copy¶

New file with same content as an existing unchanged file detected as a copy

Browse source: directory-file-copy
Tags: copy, directory, content-hash
Snapshots: snapshot-a has 1 file — original.txt; snapshot-b has 2 files — duplicate.txt, original.txt

Run it:

binoc diff \
  ./test-vectors-materialized/directory-file-copy/snapshot-a \
  ./test-vectors-materialized/directory-file-copy/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **duplicate.txt**: Copied from original.txt

directory-nested¶

Subdirectories with mixed changes

Browse source: directory-nested
Tags: directory, nested, mixed
Snapshots: snapshot-a has 2 files — data/records.csv, docs/readme.txt; snapshot-b has 3 files — data/extra.csv, data/records.csv, docs/readme.txt

Run it:

binoc diff \
  ./test-vectors-materialized/directory-nested/snapshot-a \
  ./test-vectors-materialized/directory-nested/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data/records.csv**: 1 row added
  - Rows added
    - row 3: '3', 'Charlie'
- **data/extra.csv**: Added
- **docs/readme.txt**: 2 lines added; 1 line removed
  - Line changes
    - line 1: 'Version 1 readme' -> 'Version 2 readme'

directory-nested-with-tar¶

Shows binoc diffing a tar archive and a plain directory that contain overlapping internal paths.

Browse source: directory-nested-with-tar
Tags: directory, tar, overlap, artifact-collision
Snapshots: snapshot-a has 2 files — data.tar.gz, data/records.csv; snapshot-b has 2 files — data.tar.gz, data/records.csv

Run it:

binoc diff \
  ./test-vectors-materialized/directory-nested-with-tar/snapshot-a \
  ./test-vectors-materialized/directory-nested-with-tar/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.tar.gz/>records.csv**: 1 cell changed
  - Changed cells
    - row 2, column 'count': '20' -> '25'
- **data/records.csv**: 1 row added
  - Rows added
    - row 3: '3', 'Charlie'

enforcement-actions-merge-years¶

Per-year CSVs merged row-wise into one file; detected as a clean partition merge (CFM-72)

Browse source: enforcement-actions-merge-years
Tags: csv, partition, merge
Snapshots: snapshot-a has 2 files — actions_2023.csv, actions_2024.csv; snapshot-b has 1 file — actions.csv

Run it:

binoc diff \
  ./test-vectors-materialized/enforcement-actions-merge-years/snapshot-a \
  ./test-vectors-materialized/enforcement-actions-merge-years/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

Claims

- actions_2023.csv, actions_2024.csv merged into actions.csv

- **actions.csv**: Merged from actions_2023.csv, actions_2024.csv

file-correspondence-container¶

Config declares a correspondence between renamed zip containers

Browse source: file-correspondence-container
Tags: zip, file-correspondence, declared-correspondence, container
Snapshots: snapshot-a has 1 file — data.zip; snapshot-b has 1 file — archive.zip
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/file-correspondence-container.yaml:

dataset:
  files:
    correspondences:
      - name: archive-pair
        key: archive
        left:
          path_regex: ^data\.zip$
        right:
          path_regex: ^archive\.zip$

Run it:

binoc diff \
  ./test-vectors-materialized/file-correspondence-container/snapshot-a \
  ./test-vectors-materialized/file-correspondence-container/snapshot-b \
  --config /tmp/file-correspondence-container.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.zip**: Moved from data.zip

file-correspondence-scheme¶

Config declares that a state CSV moved into a new directory scheme is the same logical file

Browse source: file-correspondence-scheme
Tags: csv, file-correspondence, scheme-change
Snapshots: snapshot-a has 1 file — data/state_AL.csv; snapshot-b has 1 file — by-state/AL/records.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/file-correspondence-scheme.yaml:

dataset:
  files:
    correspondences:
      - name: state-records
        key: "${state}"
        logical_path: "states/${state}.csv"
        on_null_key: diagnostic
        on_duplicate_key: diagnostic
        left:
          path_regex: "^data/state_(?P<state>[A-Z]{2})\\.csv$"
        right:
          path_regex: "^by-state/(?P<state>[A-Z]{2})/records\\.csv$"

Run it:

binoc diff \
  ./test-vectors-materialized/file-correspondence-scheme/snapshot-a \
  ./test-vectors-materialized/file-correspondence-scheme/snapshot-b \
  --config /tmp/file-correspondence-scheme.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **by-state**: Added
- **by-state/AL**: Moved from data
- **by-state/AL/records.csv**:
  - Moved from data/state_AL.csv
  - 1 row added
  - Rows added
    - row 2: '2', 'Birmingham'

file-correspondence-token¶

Config declares that year-stamped CSV filenames are the same logical file

Browse source: file-correspondence-token
Tags: csv, file-correspondence, declared-correspondence
Snapshots: snapshot-a has 1 file — running_list_as_of_2022.csv; snapshot-b has 1 file — running_list_as_of_2023.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/file-correspondence-token.yaml:

dataset:
  files:
    correspondences:
      - name: running-list
        key: "${list}"
        logical_path: "${list}.csv"
        on_null_key: diagnostic
        on_duplicate_key: diagnostic
        left:
          path_regex: "^(?P<list>running_list)_as_of_[0-9]{4}\\.csv$"
        right:
          path_regex: "^(?P<list>running_list)_as_of_[0-9]{4}\\.csv$"

Run it:

binoc diff \
  ./test-vectors-materialized/file-correspondence-token/snapshot-a \
  ./test-vectors-materialized/file-correspondence-token/snapshot-b \
  --config /tmp/file-correspondence-token.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **running_list_as_of_2023.csv**:
  - Moved from running_list_as_of_2022.csv
  - 1 row added
  - Rows added
    - row 3: '3', 'Cy'

folder-move-nested¶

Detects a whole-folder rename and rolls many file moves up into one folder-move entry.

Browse source: folder-move-nested
Tags: folder-move, rollup, nested, directory
Snapshots: snapshot-a has 4 files — docs/readme.txt, docs/reports/annual.txt, docs/reports/quarterly/q1.txt, docs/reports/quarterly/q2.txt; snapshot-b has 4 files — documentation/readme.txt, documentation/reports/annual.txt, documentation/reports/quarterly/q1.txt, documentation/reports/quarterly/q2.txt

Run it:

binoc diff \
  ./test-vectors-materialized/folder-move-nested/snapshot-a \
  ./test-vectors-materialized/folder-move-nested/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **documentation**: Moved from docs

folder-move-partial¶

Detects a mostly-moved folder rename and preserves only the added/removed/modified remainder entries beneath it.

Browse source: folder-move-partial
Tags: folder-move, partial, rollup, directory
Snapshots: snapshot-a has 10 files — FoodData_Central_csv_2025-12-18/README.txt, FoodData_Central_csv_2025-12-18/data/categories.csv, FoodData_Central_csv_2025-12-18/data/food.csv, FoodData_Central_csv_2025-12-18/data/nutrients.csv, +6 more; snapshot-b has 10 files — FoodData_Central_csv_2026-04-30/README.txt, FoodData_Central_csv_2026-04-30/data/categories.csv, FoodData_Central_csv_2026-04-30/data/food.csv, FoodData_Central_csv_2026-04-30/data/new-table.csv, +6 more

Run it:

binoc diff \
  ./test-vectors-materialized/folder-move-partial/snapshot-a \
  ./test-vectors-materialized/folder-move-partial/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **FoodData_Central_csv_2026-04-30**: Added
- **FoodData_Central_csv_2026-04-30/README.txt**: Moved from FoodData_Central_csv_2025-12-18/README.txt
- **FoodData_Central_csv_2026-04-30/data**: Moved from FoodData_Central_csv_2025-12-18/data
- **FoodData_Central_csv_2026-04-30/data/new-table.csv**: Added
- **FoodData_Central_csv_2026-04-30/docs**: Added
- **FoodData_Central_csv_2026-04-30/docs/changelog-note.txt**: Moved from FoodData_Central_csv_2025-12-18/docs/changelog-note.txt
- **FoodData_Central_csv_2026-04-30/docs/license.txt**: Moved from FoodData_Central_csv_2025-12-18/docs/license.txt
- **FoodData_Central_csv_2026-04-30/docs/schema.txt**: Moved from FoodData_Central_csv_2025-12-18/docs/schema.txt
- **FoodData_Central_csv_2026-04-30/docs/modified.txt**: Added
- **FoodData_Central_csv_2025-12-18**: Removed
- **FoodData_Central_csv_2025-12-18/docs**: Removed
- **FoodData_Central_csv_2025-12-18/docs/modified.txt**: Removed
- **FoodData_Central_csv_2025-12-18/docs/old-table.txt**: Removed

geojson-feature-cell-change¶

A GeoJSON FeatureCollection where one feature's property changes; transcoded to a tabular artifact with the geometry as…

Browse source: geojson-feature-cell-change
Tags: geojson, tabular, nested, cell-change
Snapshots: snapshot-a has 1 file — places.geojson; snapshot-b has 1 file — places.geojson

Run it:

binoc diff \
  ./test-vectors-materialized/geojson-feature-cell-change/snapshot-a \
  ./test-vectors-materialized/geojson-feature-cell-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **places.geojson**: 1 cell changed
  - Changed cells
    - row 1, column 'properties': {"name":"Boston","population":650000} -> {"name":"Boston","population":675000}

gzip-inner-dispatch¶

Gzipped CSV and text are decompressed and redispatched under their inner names

Browse source: gzip-inner-dispatch
Tags: gzip, csv, text, cell-change, row-addition, line-change
Snapshots: snapshot-a has 2 files — census.txt.gz, data.csv.gz; snapshot-b has 2 files — census.txt.gz, data.csv.gz

Run it:

binoc diff \
  ./test-vectors-materialized/gzip-inner-dispatch/snapshot-a \
  ./test-vectors-materialized/gzip-inner-dispatch/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **census.txt.gz/>census.txt**: 1 line added; 1 line removed
  - Line changes
    - line 2: '1|Aroostook|120' -> '1|Aroostook|121'
- **data.csv.gz/>data.csv**: 1 row added; 1 cell changed
  - Changed cells
    - row 2, column 'name': 'Bob' -> 'Robert'
  - Rows added
    - row 3: '3', 'Carla'

ini-value-change¶

An INI value changes; transcoded to a structured_document and reported as a value change

Browse source: ini-value-change
Tags: ini, structured-document, value-change
Snapshots: snapshot-a has 1 file — config.ini; snapshot-b has 1 file — config.ini

Run it:

binoc diff \
  ./test-vectors-materialized/ini-value-change/snapshot-a \
  ./test-vectors-materialized/ini-value-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **config.ini**: $.replicas: "3" -> "5"

json-array-order-significant¶

JSON array order changes are semantic content changes in stage 1

Browse source: json-array-order-significant
Tags: json, array-order, content-change
Snapshots: snapshot-a has 1 file — metadata.json; snapshot-b has 1 file — metadata.json

Run it:

binoc diff \
  ./test-vectors-materialized/json-array-order-significant/snapshot-a \
  ./test-vectors-materialized/json-array-order-significant/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **metadata.json**: $.ids[1]: 2 -> 3; $.ids[2]: 3 -> 2

json-key-order-reexport¶

JSON object key order and pretty-printing changed without semantic value changes

Browse source: json-key-order-reexport
Tags: json, serialization, key-order
Snapshots: snapshot-a has 1 file — metadata.json; snapshot-b has 1 file — metadata.json

Run it:

binoc diff \
  ./test-vectors-materialized/json-key-order-reexport/snapshot-a \
  ./test-vectors-materialized/json-key-order-reexport/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **metadata.json**: Document serialization changed

json-keyed-row-diff¶

Configured JSON record keys match reordered rows and report keyed row/cell changes

Browse source: json-keyed-row-diff
Tags: json, keyed, row-addition, row-removal, cell-change
Snapshots: snapshot-a has 1 file — data.json; snapshot-b has 1 file — data.json
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/json-keyed-row-diff.yaml:

dataset:
  paths:
    - match: data.json
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/json-keyed-row-diff/snapshot-a \
  ./test-vectors-materialized/json-keyed-row-diff/snapshot-b \
  --config /tmp/json-keyed-row-diff.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **data.json**: 1 row added; 1 row removed; 1 row modified by key
  - Changed cells
    - key id 'p2', column 'price': 20 -> 25
  - Rows added
    - key id 'p4': 'p4', 'Delta', 40
  - Rows removed
    - key id 'p3': 'p3', 'Gamma', 30

json-records-cell-change¶

JSON array of like-shaped objects parsed as a typed table; numeric cell values change

Browse source: json-records-cell-change
Tags: json, records, tabular, cell-change
Snapshots: snapshot-a has 1 file — data.json; snapshot-b has 1 file — data.json

Run it:

binoc diff \
  ./test-vectors-materialized/json-records-cell-change/snapshot-a \
  ./test-vectors-materialized/json-records-cell-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.json**: 2 cells changed
  - Changed cells
    - row 1, column 'score': 85 -> 92
    - row 2, column 'score': 90 -> 88

json-records-nested-value¶

JSON records with a nested object cell; the nested value changes and is reported as a single equality-based cell edit (…

Browse source: json-records-nested-value
Tags: json, records, tabular, nested, cell-change
Snapshots: snapshot-a has 1 file — people.json; snapshot-b has 1 file — people.json

Run it:

binoc diff \
  ./test-vectors-materialized/json-records-nested-value/snapshot-a \
  ./test-vectors-materialized/json-records-nested-value/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **people.json**: 1 cell changed
  - Changed cells
    - row 2, column 'meta': {"role":"user","tags":["z"]} -> {"role":"editor","tags":["z"]}

json-records-path-stix-objects¶

STIX-shaped JSON bundle records under $.objects are parsed as a keyed table

Browse source: json-records-path-stix-objects
Tags: json, records, records-path, stix, keyed
Snapshots: snapshot-a has 1 file — enterprise.stix.json; snapshot-b has 1 file — enterprise.stix.json
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/json-records-path-stix-objects.yaml:

dataset:
  paths:
    - match: **/*.stix.json
      records_path: $.objects
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/json-records-path-stix-objects/snapshot-a \
  ./test-vectors-materialized/json-records-path-stix-objects/snapshot-b \
  --config /tmp/json-records-path-stix-objects.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **enterprise.stix.json**: 1 row modified by key
  - Changed cells
    - key id 'attack-pattern--1', column 'name': 'Old Name' -> 'New Name'

json-records-type-only-column¶

Typed JSON records re-serialized with one numeric column as strings; canonical values match

Browse source: json-records-type-only-column
Tags: json, records, tabular, column-type-change
Snapshots: snapshot-a has 1 file — data.json; snapshot-b has 1 file — data.json

Run it:

binoc diff \
  ./test-vectors-materialized/json-records-type-only-column/snapshot-a \
  ./test-vectors-materialized/json-records-type-only-column/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.json**: Column type changed: 'year' number -> string

jsonl-row-addition¶

JSONL stream of like-shaped objects parsed as a table; a record is appended

Browse source: jsonl-row-addition
Tags: jsonl, records, tabular, row-addition
Snapshots: snapshot-a has 1 file — events.jsonl; snapshot-b has 1 file — events.jsonl

Run it:

binoc diff \
  ./test-vectors-materialized/jsonl-row-addition/snapshot-a \
  ./test-vectors-materialized/jsonl-row-addition/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **events.jsonl**: 1 row added
  - Rows added
    - row 3: 'carol', 'delete', 3

jsonld-value-change¶

A .jsonld file with no declared media type parses as a structured document tagged format=jsonld; a value change is repo…

Browse source: jsonld-value-change
Tags: json, jsonld, structured-document
Snapshots: snapshot-a has 1 file — person.jsonld; snapshot-b has 1 file — person.jsonld

Run it:

binoc diff \
  ./test-vectors-materialized/jsonld-value-change/snapshot-a \
  ./test-vectors-materialized/jsonld-value-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **person.jsonld**: $.jobTitle: "Mathematician" -> "Computer Scientist"

kitchen-sink¶

Runs text, CSV, archive, move, and copy detection together in one end-to-end example.

Browse source: kitchen-sink
Tags: csv, text, binary, tar, zip, directory, move, copy, column-reorder, integration
Snapshots: snapshot-a has 9 files — archive.tar.gz, bundle.zip, data.csv, docs/old-notes.txt, +5 more; snapshot-b has 10 files — archive.tar.gz, bundle.zip, data.csv, docs/new-file.txt, +6 more

Run it:

binoc diff \
  ./test-vectors-materialized/kitchen-sink/snapshot-a \
  ./test-vectors-materialized/kitchen-sink/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.tar.gz/>inventory.csv**: 1 row added
  - Rows added
    - row 3: 'sprockets', '20'
- **bundle.zip/>notes.txt**: 2 lines added; 1 line removed
  - Line changes
    - line 1: 'Version 1 notes.' -> 'Version 2 notes.'
- **data.csv**: 2 cells changed
  - Changed cells
    - row 1, column 'age': '30' -> '31'
    - row 3, column 'city': 'Seattle' -> 'Portland'
- **docs/readme.txt**: 2 lines added; 2 lines removed
  - Line changes
    - line 2: 'This is the original readme.' -> 'This is the updated readme.'
    - line 4: 'Some will change.' -> 'New content added here.'
- **docs/old-notes.txt**: Removed
- **docs/new-file.txt**: Added
- **icon.bin**: Binary content changed; 1 extracted string added, 1 extracted string removed
  - Extracted strings added
    - '\nFAKEICONv2'
  - Extracted strings removed
    - '\nFAKEICONv1'
- **license-copy.txt**: Copied from license.txt
- **metrics.csv**: Columns reordered
- **summary.txt**: Moved from report.txt

nasa-gistemp-header-line¶

NASA GISTEMP-style table with a title line before the real header

Browse source: nasa-gistemp-header-line
Tags: csv, config, shape, header-line, keyed
Snapshots: snapshot-a has 1 file — GLB.Ts+dSST.csv; snapshot-b has 1 file — GLB.Ts+dSST.csv
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/nasa-gistemp-header-line.yaml:

dataset:
  paths:
    - match: **/GLB.Ts+dSST.csv
      content_type: text/csv
      shape:
        header_line: 2
      row_identity:
        columns:
          - Year

Run it:

binoc diff \
  ./test-vectors-materialized/nasa-gistemp-header-line/snapshot-a \
  ./test-vectors-materialized/nasa-gistemp-header-line/snapshot-b \
  --config /tmp/nasa-gistemp-header-line.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **GLB.Ts+dSST.csv**: 1 row modified by key
  - Changed cells
    - key Year '2024', column 'Feb': '118' -> '119'

observations-repartition-equal-arity¶

Equal-arity N→M repartition: 2 tables grouped by region become 2 tables grouped by year, every row preserved exactly bu…

Browse source: observations-repartition-equal-arity
Tags: csv, partition, possible-split, equal-arity
Snapshots: snapshot-a has 2 files — observations_north.csv, observations_south.csv; snapshot-b has 2 files — observations_2024.csv, observations_2025.csv

Run it:

binoc diff \
  ./test-vectors-materialized/observations-repartition-equal-arity/snapshot-a \
  ./test-vectors-materialized/observations-repartition-equal-arity/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **observations_2024.csv**:
  - Moved from observations_north.csv
  - 3 cells changed
  - Changed cells
    - row 2, column 'year': '2025' -> '2024'
    - row 2, column 'region': 'north' -> 'south'
    - row 2, column 'count': '15' -> '12'
- **observations_2025.csv**:
  - Moved from observations_south.csv
  - 3 cells changed
  - Changed cells
    - row 1, column 'year': '2024' -> '2025'
    - row 1, column 'region': 'south' -> 'north'
    - row 1, column 'count': '12' -> '15'

## Suggestions

- 'observations_north.csv' shares rows with other unmatched tables but the relationship is not a clean partition (residual, shared, or extra rows); left as add/remove (`binoc.pair.partition`) [binoc.possible_split]

observations-split-by-year¶

One CSV split row-wise into per-year files; detected as a clean partition split (CFM-72)

Browse source: observations-split-by-year
Tags: csv, partition, split
Snapshots: snapshot-a has 1 file — observations.csv; snapshot-b has 2 files — observations_2024.csv, observations_2025.csv

Run it:

binoc diff \
  ./test-vectors-materialized/observations-split-by-year/snapshot-a \
  ./test-vectors-materialized/observations-split-by-year/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

Claims

- observations.csv split into observations_2024.csv, observations_2025.csv

- **observations_2024.csv**: Split from observations.csv
- **observations_2025.csv**: Split from observations.csv

observations-split-residual¶

A would-be split missing one row: partition declines (not complete), emits binoc.possible_split, and degrades to honest…

Browse source: observations-split-residual
Tags: csv, partition, possible-split
Snapshots: snapshot-a has 1 file — observations.csv; snapshot-b has 2 files — observations_2024.csv, observations_2025.csv

Run it:

binoc diff \
  ./test-vectors-materialized/observations-split-residual/snapshot-a \
  ./test-vectors-materialized/observations-split-residual/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **observations_2024.csv**:
  - Moved from observations.csv
  - 2 rows removed
  - Rows removed
    - row 3: '2025', 'north', '15'
    - row 4: '2025', 'south', '9'
- **observations_2025.csv**: Added

## Suggestions

- 'observations.csv' shares rows with other unmatched tables but the relationship is not a clean partition (residual, shared, or extra rows); left as add/remove (`binoc.pair.partition`) [binoc.possible_split]

ofac-sdn-headerless-position-key¶

OFAC SDN-style 12-column headerless CSV keyed by ent_num in column 1

Browse source: ofac-sdn-headerless-position-key
Tags: csv, config, shape, headerless, keyed
Snapshots: snapshot-a has 1 file — SDN.CSV; snapshot-b has 1 file — SDN.CSV
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/ofac-sdn-headerless-position-key.yaml:

dataset:
  paths:
    - match: **/SDN.CSV
      content_type: text/csv
      shape:
        has_header: false
      row_identity:
        by_position:
          - 1

Run it:

binoc diff \
  ./test-vectors-materialized/ofac-sdn-headerless-position-key/snapshot-a \
  ./test-vectors-materialized/ofac-sdn-headerless-position-key/snapshot-b \
  --config /tmp/ofac-sdn-headerless-position-key.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **SDN.CSV**: 1 row modified by key
  - Changed cells
    - key column_1 '100', column 'column_11': 'remarks' -> 'remarks-updated'

per-path-dispatch-override¶

Per-path dispatch overrides promote extensionless CSV content before tabular row keying

Browse source: per-path-dispatch-override
Tags: config, dispatch, csv, keyed
Snapshots: snapshot-a has 2 files — forced, records; snapshot-b has 2 files — forced, records
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/per-path-dispatch-override.yaml:

dataset:
  defaults:
    row_identity:
      columns:
        - id
  paths:
    - match: **/records
      content_type: text/csv
    - match: forced
      rule: binoc.parse.csv
      row_identity:
        columns:
          - id

Run it:

binoc diff \
  ./test-vectors-materialized/per-path-dispatch-override/snapshot-a \
  ./test-vectors-materialized/per-path-dispatch-override/snapshot-b \
  --config /tmp/per-path-dispatch-override.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **forced**: 1 row modified by key
  - Changed cells
    - key id '1', column 'value': 'old' -> 'new'
- **records**: 1 row modified by key
  - Changed cells
    - key id '1', column 'value': 'old' -> 'new'

single-file-add¶

File present in B but not A

Browse source: single-file-add
Tags: add, file
Snapshots: snapshot-a has 0 files (empty snapshot); snapshot-b has 1 file — new_file.txt

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-add/snapshot-a \
  ./test-vectors-materialized/single-file-add/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **new_file.txt**: Added

single-file-modify-binary¶

Binary file, different hash

Browse source: single-file-modify-binary
Tags: modify, binary
Snapshots: snapshot-a has 1 file — data.bin; snapshot-b has 1 file — data.bin

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-modify-binary/snapshot-a \
  ./test-vectors-materialized/single-file-modify-binary/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.bin**: 1 edit

single-file-modify-csv¶

CSV file compared directly (file-to-file, not via directory)

Browse source: single-file-modify-csv
Tags: csv, single-file, modify
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-modify-csv/snapshot-a/data.csv \
  ./test-vectors-materialized/single-file-modify-csv/snapshot-b/data.csv

Result:

# Changelog: snapshot-a → snapshot-b

- **data.csv**: 1 row added
  - Rows added
    - row 3: 'Charlie', '35'

single-file-modify-text¶

Text file with line-level changes

Browse source: single-file-modify-text
Tags: modify, text, lines
Snapshots: snapshot-a has 1 file — story.txt; snapshot-b has 1 file — story.txt

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-modify-text/snapshot-a \
  ./test-vectors-materialized/single-file-modify-text/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **story.txt**: 2 lines added; 1 line removed
  - Line changes
    - line 2: 'Line 2' -> 'Line 2 revised'

single-file-modify-text-root¶

Text file compared directly (file-to-file, not via directory)

Browse source: single-file-modify-text-root
Tags: text, single-file, modify
Snapshots: snapshot-a has 1 file — story.txt; snapshot-b has 1 file — story.txt

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-modify-text-root/snapshot-a/story.txt \
  ./test-vectors-materialized/single-file-modify-text-root/snapshot-b/story.txt

Result:

# Changelog: snapshot-a → snapshot-b

- **story.txt**: 2 lines added; 1 line removed
  - Line changes
    - line 2: 'Line 2' -> 'Line 2 revised'

single-file-remove¶

File present in A but not B

Browse source: single-file-remove
Tags: remove, file
Snapshots: snapshot-a has 1 file — removed_file.txt; snapshot-b has 0 files (empty snapshot)

Run it:

binoc diff \
  ./test-vectors-materialized/single-file-remove/snapshot-a \
  ./test-vectors-materialized/single-file-remove/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **removed_file.txt**: Removed

stacked-csv-broken-out¶

Stacked-CSV tables broken out into one file per table; whole-table rehoming (reshape + 1:1), NOT a partition split (CFM…

Browse source: stacked-csv-broken-out
Tags: csv, stacked-tables, reshape
Snapshots: snapshot-a has 1 file — report.csv; snapshot-b has 2 files — changes.csv, products.csv

Run it:

binoc diff \
  ./test-vectors-materialized/stacked-csv-broken-out/snapshot-a \
  ./test-vectors-materialized/stacked-csv-broken-out/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **changes.csv**: Moved from report.csv/>table_1
- **products.csv**: Reshaped from report.csv (stacked tables → tabular)
- **report.csv/>table_2**: Removed

tar-nested¶

Nested tar.gz containing CSV

Browse source: tar-nested
Tags: tar, nested, csv
Snapshots: snapshot-a has 1 file — outer.tar.gz; snapshot-b has 1 file — outer.tar.gz

Run it:

binoc diff \
  ./test-vectors-materialized/tar-nested/snapshot-a \
  ./test-vectors-materialized/tar-nested/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **outer.tar.gz/>inner.tar.gz/>data.csv**: 1 row added
  - Rows added
    - row 2: 'Bob', '25'

tar-simple¶

Tar.gz archive with changes inside

Browse source: tar-simple
Tags: tar, archive
Snapshots: snapshot-a has 1 file — archive.tar.gz; snapshot-b has 1 file — archive.tar.gz

Run it:

binoc diff \
  ./test-vectors-materialized/tar-simple/snapshot-a \
  ./test-vectors-materialized/tar-simple/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.tar.gz/>data.csv**: 1 row added
  - Rows added
    - row 3: 'gamma', '3'
- **archive.tar.gz/>hello.txt**: 1 line added

text-rename-modify¶

Text file renamed and modified: detected as a single move by fuzzy correlation

Browse source: text-rename-modify
Tags: text, fuzzy-move, rename-modify
Snapshots: snapshot-a has 1 file — notes.txt; snapshot-b has 1 file — meeting-notes-v2.txt

Run it:

binoc diff \
  ./test-vectors-materialized/text-rename-modify/snapshot-a \
  ./test-vectors-materialized/text-rename-modify/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **meeting-notes-v2.txt**:
  - Moved from notes.txt
  - 2 lines added
  - Line changes (showing 3 of 4)
    - line 10: '' -> '- Marketing strategy update'
    - line 11: 'Action Items:' -> ''
    - line 12: '- Alice to finalize budget by Friday' -> 'Action Items:'

toml-value-change¶

A TOML value changes; transcoded to a structured_document and reported as a value change

Browse source: toml-value-change
Tags: toml, structured-document, value-change
Snapshots: snapshot-a has 1 file — config.toml; snapshot-b has 1 file — config.toml

Run it:

binoc diff \
  ./test-vectors-materialized/toml-value-change/snapshot-a \
  ./test-vectors-materialized/toml-value-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **config.toml**: $.replicas: 3 -> 5

tree-wide-correlation¶

Shows tree-wide move and copy detection across nested zip boundaries, including one-to-many copies and many-to-one moves.

Browse source: tree-wide-correlation
Tags: move, copy, aggregation, zip, nested, archive, tree-wide
Snapshots: snapshot-a has 4 files — alpha.txt, dup.bin, kept.txt, outer.zip; snapshot-b has 5 files — gamma-renamed.txt, kept-copy.txt, kept.txt, merged.bin, +1 more

Run it:

binoc diff \
  ./test-vectors-materialized/tree-wide-correlation/snapshot-a \
  ./test-vectors-materialized/tree-wide-correlation/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **gamma-renamed.txt**: Moved from outer.zip/>inner.zip/>gamma.txt
- **kept-copy.txt**: Copied from kept.txt
- **merged.bin**: Moved from dup.bin
- **outer.zip/>alpha-renamed.txt**: Moved from alpha.txt
- **outer.zip/>inner.zip/>beta-renamed.txt**: Moved from outer.zip/>beta.txt
- **outer.zip/>kept-copy.txt**: Copied from kept.txt
- **outer.zip/>dup-b.bin**: Removed

trivial-identical¶

Two identical directories → empty changeset

Browse source: trivial-identical
Tags: identical, baseline
Snapshots: snapshot-a has 1 file — data.txt; snapshot-b has 1 file — data.txt

Run it:

binoc diff \
  ./test-vectors-materialized/trivial-identical/snapshot-a \
  ./test-vectors-materialized/trivial-identical/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

trivial-identical-csv¶

Two identical CSV files → no changes reported

Browse source: trivial-identical-csv
Tags: csv, identical, baseline
Snapshots: snapshot-a has 1 file — data.csv; snapshot-b has 1 file — data.csv

Run it:

binoc diff \
  ./test-vectors-materialized/trivial-identical-csv/snapshot-a \
  ./test-vectors-materialized/trivial-identical-csv/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

tsv-cell-changes¶

Tab-delimited file parses into real columns and reports cell changes

Browse source: tsv-cell-changes
Tags: tsv, cell-change
Snapshots: snapshot-a has 1 file — data.tsv; snapshot-b has 1 file — data.tsv

Run it:

binoc diff \
  ./test-vectors-materialized/tsv-cell-changes/snapshot-a \
  ./test-vectors-materialized/tsv-cell-changes/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.tsv**: 2 cells changed
  - Dialect provenance: detected tab-delimited, no quoting, newline LF
  - Changed cells
    - row 1, column 'age': '30' -> '31'
    - row 2, column 'city': 'Boston' -> 'Cambridge'

yaml-value-change¶

A YAML scalar value changes; transcoded to a structured_document and reported as a value change

Browse source: yaml-value-change
Tags: yaml, structured-document, value-change
Snapshots: snapshot-a has 1 file — config.yaml; snapshot-b has 1 file — config.yaml

Run it:

binoc diff \
  ./test-vectors-materialized/yaml-value-change/snapshot-a \
  ./test-vectors-materialized/yaml-value-change/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **config.yaml**: $.replicas: 3 -> 5

zip-declared-container¶

Config declares a correspondence between nested zip containers and preserves inner CSV content detail

Browse source: zip-declared-container
Tags: zip, file-correspondence, declared-correspondence, container
Snapshots: snapshot-a has 1 file — outer.zip; snapshot-b has 1 file — outer.zip
Setup: This example uses a custom dataset config to make the relevant correspondence behavior obvious. Save this dataset config as /tmp/zip-declared-container.yaml:

dataset:
  files:
    correspondences:
      - name: inner-archive-pair
        key: records
        logical_path: outer.zip/>records.zip
        on_null_key: diagnostic
        on_duplicate_key: diagnostic
        left:
          path_regex: ^outer\.zip/>records-old\.zip$
        right:
          path_regex: ^outer\.zip/>records\.zip$

Run it:

binoc diff \
  ./test-vectors-materialized/zip-declared-container/snapshot-a \
  ./test-vectors-materialized/zip-declared-container/snapshot-b \
  --config /tmp/zip-declared-container.yaml

Result:

# Changelog: snapshot-a → snapshot-b

- **outer.zip/>records.zip**: Moved from outer.zip/>records-old.zip
- **outer.zip/>records.zip/>data.csv**: 1 cell changed
  - Changed cells
    - row 2, column 'score': '24' -> '25'

zip-json-key-order-reexport¶

JSON files inside zip expansion get parsed and rendered as serialization-only changes

Browse source: zip-json-key-order-reexport
Tags: zip, json, serialization, key-order
Snapshots: snapshot-a has 1 file — archive.zip; snapshot-b has 1 file — archive.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-json-key-order-reexport/snapshot-a \
  ./test-vectors-materialized/zip-json-key-order-reexport/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.zip/>metadata.json**: Document serialization changed

zip-nested¶

Nested zip containing CSV

Browse source: zip-nested
Tags: zip, nested, csv
Snapshots: snapshot-a has 1 file — outer.zip; snapshot-b has 1 file — outer.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-nested/snapshot-a \
  ./test-vectors-materialized/zip-nested/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **outer.zip/>inner.zip/>data.csv**: 1 row added
  - Rows added
    - row 2: 'Bob', '25'

zip-rename-contents-rewritten¶

Documents a known gap — a renamed zip whose children were all renamed AND rewritten (no content similarity) yields unpa…

Browse source: zip-rename-contents-rewritten
Tags: zip, archive, known-gap
Snapshots: snapshot-a has 1 file — data.zip; snapshot-b has 1 file — archive.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-rename-contents-rewritten/snapshot-a \
  ./test-vectors-materialized/zip-rename-contents-rewritten/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **data.zip**: Removed
- **data.zip/>x.csv**: Removed
- **data.zip/>y.csv**: Removed
- **data.zip/>z.csv**: Removed
- **archive.zip**: Added
- **archive.zip/>p.csv**: Added
- **archive.zip/>q.csv**: Added
- **archive.zip/>r.csv**: Added

zip-rename-identical¶

Zip archive renamed with identical contents; bottom-up roll-up of the inner clean file moves compacts the pair into a s…

Browse source: zip-rename-identical
Tags: zip, archive, folder-move
Snapshots: snapshot-a has 1 file — data.zip; snapshot-b has 1 file — archive.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-rename-identical/snapshot-a \
  ./test-vectors-materialized/zip-rename-identical/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.zip**: Moved from data.zip

zip-rename-inner-rename-edit¶

Zip archive renamed while its only child was renamed and had one cell edited; the modified move counts as roll-up evide…

Browse source: zip-rename-inner-rename-edit
Tags: zip, archive, folder-move, fuzzy-correlation
Snapshots: snapshot-a has 1 file — data.zip; snapshot-b has 1 file — archive.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-rename-inner-rename-edit/snapshot-a \
  ./test-vectors-materialized/zip-rename-inner-rename-edit/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.zip**: Moved from data.zip
- **archive.zip/>new.csv**:
  - Moved from data.zip/>old.csv
  - 1 cell changed
  - Changed cells
    - row 5, column 'score': '60' -> '61'

zip-simple¶

Zipped files with changes inside

Browse source: zip-simple
Tags: zip, archive
Snapshots: snapshot-a has 1 file — archive.zip; snapshot-b has 1 file — archive.zip

Run it:

binoc diff \
  ./test-vectors-materialized/zip-simple/snapshot-a \
  ./test-vectors-materialized/zip-simple/snapshot-b

Result:

# Changelog: snapshot-a → snapshot-b

- **archive.zip/>data.txt**: 1 line added; 1 line removed
  - Line changes
    - line 1: 'hello from zip A' -> 'hello from zip B'
- **archive.zip/>extra.txt**: Added