FAILURE MAP
← Case archive

FA-79246 / Image orientation metadata / Open access

Metadata stripping counts only for baked uploads · case 01

Stripped but unbaked uploads are predicted to rotate even though no tag survives.

Verified by executionVariant 1 · 8 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The remaining tag is cleared for stripped files only when they were also baked.

THE FAILURE

The remaining tag is cleared for stripped files only when they were also baked.

Unsuccessful approach: Clearing only unbaked stripped files keeps a stale tag on baked stripped ones.

Case contract

Predict what a browser shows for a photo that went through an upload pipeline. Input {tag, baked, reset, stripped, css, cross_origin}. Baking applies the tag to the pixels; the remaining tag is 1 if metadata was stripped or if baking also reset the tag, else the original tag. The renderer applies the remaining tag when css is "from-image" or the image is cross-origin (image-orientation: none is ignored cross-origin). Finally the viewer applies its own extra orientation (default 1). Orientations are (k clockwise quarter turns after m mirrors), with applying B after A giving k = k2 + (-k1 if m2 else k1). Invalid tags count as 1. Return [shown_tag, shown == tag].

Why this case matters

Camera, phone and scanner images carry an orientation hint separately from the stored pixels; galleries, thumbnailers, editors and upload pipelines must interpret it consistently or photos appear sideways, mirrored or doubly rotated.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(p):
    to_km = {1: (0, 0), 2: (0, 1), 3: (2, 0), 4: (2, 1), 5: (3, 1), 6: (1, 0), 7: (1, 1), 8: (3, 0)}
    from_km = {v: t for t, v in to_km.items()}
    def compose(first, second):
        k1, m1 = to_km[first]
        k2, m2 = to_km[second]
        k = (k2 + (-k1 if m2 else k1)) % 4
        return from_km[(k, m1 ^ m2)]
    tag = p['tag'] if p['tag'] in to_km else 1
    pixels = tag if p['baked'] else 1
    if p['stripped'] and p['baked']:
        meta = 1
    elif p['baked'] and p['reset']:
        meta = 1
    else:
        meta = tag
    honored = p['css'] == 'from-image' or p['cross_origin']
    shown = compose(compose(pixels, meta if honored else 1), p.get('extra', 1))
    return [shown, shown == tag]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[[{'tag': 2, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [5, True]], [{'tag': 7, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 3}, [5, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 6}, [6, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [2, True]], [{'tag': 1, 'baked': False, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 8}, [8, False]], [{'tag': 5, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [8, False]]], [[{'tag': 4, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]], [{'tag': 3, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 2}, [4, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 7}, [4, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [2, False]], [{'tag': 5, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 6, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 8}, [1, False]], [{'tag': 0, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, True]], [{'tag': 4, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]]], [[{'tag': 5, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 5}, [5, True]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': True, 'css': 'none', 'cross_origin': True, 'extra': 5}, [1, False]], [{'tag': 4, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 6}, [5, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 8}, [8, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [5, True]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [1, False]], [{'tag': 4, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 3}, [2, False]], [{'tag': 5, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [1, False]]], [[{'tag': 8, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 7}, [7, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 7}, [6, False]], [{'tag': 7, 'baked': False, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 2}, [2, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'none', 'cross_origin': False, 'extra': 6}, [7, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [8, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 1}, [8, True]], [{'tag': 6, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 5}, [7, False]], [{'tag': 3, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [1, False]]], [[{'tag': 6, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 2}, [2, False]], [{'tag': 3, 'baked': True, 'reset': False, 'stripped': True, 'css': 'none', 'cross_origin': True, 'extra': 5}, [7, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [5, False]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 0, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [5, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': True, 'extra': 1}, [5, True]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 1}, [5, True]], [{'tag': 8, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 5}, [5, False]]]]
labels = ["regression: stripped metadata", "repair trap", "combined fault", "control", "control", "boundary", "boundary", "control"]
for i, (args, expected) in enumerate(fixtures[N-1]):
    check("%s %d" % (labels[i % len(labels)], i), solve(args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression: stripped metadata 0[2, True][1, False]Failed
repair trap 1[5, True][5, True]Passed
combined fault 2[5, False][5, False]Passed
control 3[6, False][6, False]Passed
control 4[2, False][2, False]Passed
boundary 5[2, True][2, True]Passed
boundary 6[8, False][8, False]Passed
control 7[4, False][8, False]Failed

SHA-256 / 5f060ec0ab4897988afde63f70cb16dd9b585569b6abcb5062986665145dc254

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(p):
    to_km = {1: (0, 0), 2: (0, 1), 3: (2, 0), 4: (2, 1), 5: (3, 1), 6: (1, 0), 7: (1, 1), 8: (3, 0)}
    from_km = {v: t for t, v in to_km.items()}
    def compose(first, second):
        k1, m1 = to_km[first]
        k2, m2 = to_km[second]
        k = (k2 + (-k1 if m2 else k1)) % 4
        return from_km[(k, m1 ^ m2)]
    tag = p['tag'] if p['tag'] in to_km else 1
    pixels = tag if p['baked'] else 1
    if p['stripped'] and not p['baked']:
        meta = 1
    elif p['baked'] and p['reset']:
        meta = 1
    else:
        meta = tag
    honored = p['css'] == 'from-image' or p['cross_origin']
    shown = compose(compose(pixels, meta if honored else 1), p.get('extra', 1))
    return [shown, shown == tag]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[[{'tag': 2, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [5, True]], [{'tag': 7, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 3}, [5, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 6}, [6, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [2, True]], [{'tag': 1, 'baked': False, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 8}, [8, False]], [{'tag': 5, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [8, False]]], [[{'tag': 4, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]], [{'tag': 3, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 2}, [4, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 7}, [4, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [2, False]], [{'tag': 5, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 6, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 8}, [1, False]], [{'tag': 0, 'baked': False, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, True]], [{'tag': 4, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 1}, [1, False]]], [[{'tag': 5, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 5}, [5, True]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': True, 'css': 'none', 'cross_origin': True, 'extra': 5}, [1, False]], [{'tag': 4, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 6}, [5, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 8}, [8, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [5, True]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [1, False]], [{'tag': 4, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 3}, [2, False]], [{'tag': 5, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [1, False]]], [[{'tag': 8, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 7}, [7, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 7}, [6, False]], [{'tag': 7, 'baked': False, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 2}, [2, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'none', 'cross_origin': False, 'extra': 6}, [7, False]], [{'tag': 2, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [8, False]], [{'tag': 8, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 1}, [8, True]], [{'tag': 6, 'baked': True, 'reset': False, 'stripped': False, 'css': 'from-image', 'cross_origin': False, 'extra': 5}, [7, False]], [{'tag': 3, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 1}, [1, False]]], [[{'tag': 6, 'baked': False, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 2}, [2, False]], [{'tag': 3, 'baked': True, 'reset': False, 'stripped': True, 'css': 'none', 'cross_origin': True, 'extra': 5}, [7, False]], [{'tag': 2, 'baked': True, 'reset': True, 'stripped': True, 'css': 'from-image', 'cross_origin': False, 'extra': 8}, [5, False]], [{'tag': 5, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 6}, [2, False]], [{'tag': 0, 'baked': True, 'reset': False, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 5}, [5, False]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': True, 'extra': 1}, [5, True]], [{'tag': 5, 'baked': True, 'reset': True, 'stripped': False, 'css': 'none', 'cross_origin': False, 'extra': 1}, [5, True]], [{'tag': 8, 'baked': False, 'reset': False, 'stripped': True, 'css': 'from-image', 'cross_origin': True, 'extra': 5}, [5, False]]]]
labels = ["regression: stripped metadata", "repair trap", "combined fault", "control", "control", "boundary", "boundary", "control"]
for i, (args, expected) in enumerate(fixtures[N-1]):
    check("%s %d" % (labels[i % len(labels)], i), solve(args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression: stripped metadata 0[1, False][1, False]Passed
repair trap 1[1, False][5, True]Failed
combined fault 2[5, False][5, False]Passed
control 3[6, False][6, False]Passed
control 4[2, False][2, False]Passed
boundary 5[2, True][2, True]Passed
boundary 6[8, False][8, False]Passed
control 7[8, False][8, False]Passed

SHA-256 / b84e796167e705d091ba81fb7f79447d42bea4ce6fed647c6cf1416352117c49

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 8 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Sign in to the archive ↗

Verification & scope

A deterministic bounded teaching model with a stipulated contract; it makes no claim of conformance to any published specification. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:49:42.783419+00:00.

Case digest / cfb35e7a7d6c9ead41792018757dafccbc660fd100c7166305d24571a24d7c32