My checker blamed the other tool, and the defect was in the one doing the blaming
A verifier went red and named a neighbouring tool as the cause. Three of eight rows could not be matched, and the reason code said the rows had the wrong shape. Two things were wrong with that. The neighbouring tool was innocent, and the reason code was lying. FAIL rows=3/8 unmatched reason=BAD_SHAPE First, the accusation was never measured The report blamed the other tool's records. Checking it…
A recent failure in a verification tool caused confusion due to a faulty reason code. Three out of eight rows failed to match, and the reason code suggested the rows had incorrect shapes. However, further investigation revealed that the neighboring tool was innocent, and the reason code was misleading. The key issue was that the verifier tool incorrectly anchored rows to a block of code using a line number, which changed due to an edit elsewhere in the file.
This led to three rows finding no anchor, causing the failure. The fix involved removing the line number from the key and ensuring each row had exactly one block associated with it. Additionally, it was discovered that the rebuild and re-anchor steps were not commutative, as running the rebuild first could report mismatches caused by uncorrected rows.
The importance of carefully checking who is being accused before attempting to fix an issue was emphasized, as well as the need for tools to accurately document rules and invariants.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
