Repository navigation
Conversation
Two packages read sheets 7.B3 and 7.B7 of the section 7.B workbook that ssa-ssi-table-7b1-2024 pins: - ssa-ssi-table-7b3-2024 emits December 2024 recipients of federal SSI and of federally administered state supplementation, with their average monthly payments, for All areas, the 50 states and DC (128 facts). - ssa-ssi-table-7b7-2024 emits calendar 2024 federal SSI and state supplementation payments (65 facts). It leaves out the Total column, which repeats Table 7.B1's payment totals. The label register lists the 52 area values the packages share, and the 7.B1 manifest's note names all three packages that read the workbook. tests/test_source_ssa_ssi_tables_7b3_7b7.py checks each fact against the HTML capture of the same tables, the tables' sums, and Table 7.B1. The default bundle gains 193 facts, two packages and one duplicate semantic key: Table 7.B3's national federal count repeats SSI Monthly Statistics Table 1. Refs PolicyEngine/microcosm#1558. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The bundle test still pinned 443 source tables. The two packages add one each, so the pin moves to 445, and the Chronicle bundle job would fail without it. Two tests in the new module pin metadata that no fact key hashes: each record set's period and coverage (month 2024-12 for Table 7.B3, calendar 2024 for Table 7.B7), and each fact's record kind, measure label and concept relation. For example, a February start date on Table 7.B7's coverage passed every earlier test and fails the new one. Reword Table 7.B7's coverage note to "Payments during calendar year 2024.", since the federal-only record set shares it. In the manifest, say why none of the three packages emits the Northern Mariana Islands row: all three keep the Table 7.B1 package's area set. Also say that the workbook stores Table 7.B3's footnoted national state supplementation cells as text, and name all three tables the HTML capture renders. No fact key hashes period coverage or manifest notes, so no key moves. The changelog fragment now names the national figures Table 7.B3 leaves out. Refs PolicyEngine/microcosm#1558. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
scripts/build_ssa_ssi_7b_packages.py writes both package YAMLs from SSA's workbook and the 7.B1 package's geography rows. Its --check flag compares its output with the committed files, and a new test fails if a committed YAML drifts from what the script writes. Refs PolicyEngine/microcosm#1558. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The module pinned only the semantic-key digest. The semantic key hashes no source release, source series or observed measure, so a wrong source table, vintage or source concept passed every test in the module. test_aggregate_identity_digest pins the sorted (source_record_id, aggregate_fact_key) lines, and each of those three edits now fails it. The changelog now says that Table 7.B7's package leaves out the Total column, which repeats Table 7.B1's payment totals. The generator and test docstrings now say that the 7.B1 package emits those totals for every row but the Northern Mariana Islands', and that the generator reads row labels from the workbook. The measure labels come from the generator itself. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
test_table_7b7_total_column_repeats_table_7b1_payment_totals now builds the Table 7.B1 package and checks its calendar 2024 total-payment facts: one each for All areas, the 50 states and DC, equal to Table 7.B7's Total column times 1,000. Its docstring already said so, and nothing checked it. Five edits to the 7.B1 package's facts fail the test: dropping one total, adding $1,000 to one, duplicating one, moving one to 2023 and giving one the aged filter. The comment on AGGREGATE_IDENTITY_DIGEST drops "also". The aggregate key hashes each measure's source concept in place of its canonical concept, which only the semantic pin covers, so the module needs both pins. The generator's docstring now separates what its constants set (each measure's column and each sheet's heading and body rows) from what the script reads from the workbook's cells. bc9b650's message says a wrong source table, vintage or source concept passed every test in the module. That held when the edit went through the generator; a hand edit of the YAML alone already failed test_generator_reproduces_the_committed_packages. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
changed the base branch from
fix/ssi-table-7b1-2024-values
to
main
November 10, 2026 14:20
The Table 7.B7 test compares the 7.B1 package's totals with 7.B7's Total column. It picked the totals by their filters alone, so a total that gained a category constraint with no filter still passed. It now checks the constraints too, and it fails when a copy of the 7.B1 YAML gives Alabama's payment total the aged constraint. The generator's docstring said the script reads every row number from the workbook. The script reads each area's row from the cells, while its constants set each sheet's body, heading and column-group rows, and the docstring now says so. In the identity comment, "The aggregate key" replaces an "It" that could have meant either key. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
WIP. This PR adds SSA Annual Statistical Supplement 2025 Tables 7.B3 and 7.B7 as two Chronicle packages, read from the section 7.B workbook that #294 pins, for PolicyEngine/microcosm#1558:
ssa-ssi-table-7b3-2024: December 2024 recipients of federal SSI and of federally administered state supplementation, with their average monthly payments, for All areas, the 50 states and DC (128 facts).ssa-ssi-table-7b7-2024: calendar 2024 federal SSI and state supplementation payments (65 facts).The plan before review: the changelog fragment, the judge record, CI, and the full description with the invariants the new test module checks. The PR stacks on #294 and targets its branch, and it moves to
mainwhen #294 merges.🤖 Generated with Claude Code