| ## Background | |
| This is level 1 of 5. The records for this level are in `workspace/t1/`. | |
| ## Task | |
| Hi, the DQC approved v30.0.5 last Friday and engineering would like to swap it in for the v30.0.3 rulesets we ship. Before I sign off the swap I need the regression from you. | |
| First, which rule element ids behave differently between those releases, for each supported taxonomy year, 2023 through 2026, and what changed? Kwame's diff note is in engineering mail. | |
| Second, run the DQC's published test cases for those rules that use SEC filings on our production engine build and tell me whether v30.0.5 reproduces the expected results. | |
| Third, across our regression corpus and the filings in the published test suite, which findings does either release give that the other does not? | |
| Please put v30.0.5-regression.xlsx and a short note-to-mara.txt with your go/no-go reasoning in workspace/t1/results/, following our validation reporting conventions, as for everything you send me on this. | |
| Mara | |
| ## Input | |
| - `workspace/t1/`: the records supplied with the work (`dqc_us_rules-v30.0.3/`, `dqc_us_rules/`, `edgar/`, `manuals/`, `shared/`, `taxonomies/`, `vendor/`). | |
| ## Output | |
| - `workspace/t1/results/v30.0.5-regression.xlsx` | |
| - `workspace/t1/results/note-to-mara.txt` | |