Jenner compatibility tests for 7 scripts

[lwsinclair]

Your SAS code also runs unmodified in the Jenner Analytics system. Jenner is a complete SAS-compatible system and collaborative workspace at jenneranalytics.com. I cloned this repository, ran seven of the scripts through Jenner without rewriting any of the SAS, and committed the results as a separate jenner-check/ directory so you can re-run them whenever you want a second opinion on the analysis.

This PR adds:

jenner-check/
├── README.md           Quick orientation
├── run_jenner.sh       Mac/Linux wrapper (bash + curl)
├── run_jenner.bat      Windows equivalent
├── run_jenner.sas      Pure-SAS version (PROC HTTP, no curl needed)
├── t001_data_import_csv/         Customers import + Age + Age_Range + PROC FREQ
├── t002_merge_sort_invoices/     Basket/Invoice merge + PROC SQL count
├── t003_customer_format_sql/     Custom $age_group format + stores_visits SQL
├── t004_sales_returns_split/     Single DATA step splitting Invoice by Operation
├── t005_proc_means_basket_value/ Invoice_Total_Value via PROC MEANS (CLASS/VAR/OUT)
├── t006_supplier_freq_weight/    Supplier_id from SKU + PROC FREQ WEIGHT Quantity
└── t007_rfm_proc_sql/            RFM recency/frequency/monetary table

Each tNNN_*/ directory carries the script that ran, a tiny CSV sample carved out of the matching file in Data/, the captured log and listing from a passing run, and a meta.json recording exactly what was adapted from Code/*.sas. The adaptations are limited to the FILENAME / LIBNAME plumbing (/home/u62678062/sasuser.v94/SAS_Project/... becomes a relative input/Customers.csv) — the analytic logic, PROC options, and output semantics match the source files.

How to re-run

cd jenner-check
./run_jenner.sh --all            # run all seven, summary at end
./run_jenner.sh t005_proc_means_basket_value  # run one

There is also run_jenner.sas if you'd rather run from inside SAS Studio (PROC HTTP, no curl needed), and you can paste any of the scripts straight into the hosted workspace at jenneranalytics.com.

What I noticed reading your code

You took the time to build a real custom character format ($age_group) so the age-range groups sort in the order that makes sense for the analysis rather than alphabetically — that kind of detail makes the downstream reports much easier to read, and it shows up consistently across the three views in 2_Customer_Profiling.sas. The Invoice_Total_Value pipeline in 1_Data_Pre-Processing.sas is also nicely structured: sort, merge, compute the discounted line value in one expression, then collapse to per-invoice totals with a single PROC MEANS — that's how it should be done. And tying the whole project together with the RFM table in script 6 leading into the cluster-conditioned market-basket analysis in script 8 is a clean arc for a milestone.

What to do with this PR

Merging is welcome but not expected. Closing it without comment is just as fine — the bundles live in the fork either way and there's no follow-up on my side. No response needed.

If you'd rather not see anything like this again, put no-more-prs in any comment on this PR, or open an issue titled jenner-check: opt out, and that's it.