InterSystems Programming Challenge 1: Code Golf

Minimal-code solution to the challenge: identify Gaia DR3 sources whose BP or RP flux changed by more than 100% over the observation period and write the result CSV. Optimized for the smallest possible source — both fewest lines and fewest characters (see the sibling repo gaia-benchmark for the speed-optimized, parallel variant).

The entire program is one line of Embedded Python — 516 characters — living in a single ObjectScript class method. It reads the 20 gzipped input files, extracts and filters the flux arrays, computes the percentage change, and writes the sorted CSV, without ever leaving that one expression.

Build & run

docker-compose up --build -d
docker-compose exec iris iris session iris
USER>do ^RunScript

This writes data/out/challenge_output.csv (≈10 seconds).

How it works

do ^RunScript (src/RunScript.mac) calls Gaia.Flux.Go() (src/Gaia/Flux.cls, a single line of Embedded Python) which, in one pass:

• reads each data/in/EpochPhotometry_*.csv.gz,
• for the bp_flux and rp_flux arrays of every source, keeps only the valid fluxes and takes their min / max,
• computes ((max_flux - min_flux) / min_flux) * 100 per band and keeps the larger of the BP and RP values as percentage_change,
• writes the sources with percentage_change > 100 (sorted descending) to the CSV.

Output columns: source_id, bp_min_flux, bp_max_flux, rp_min_flux, rp_max_flux, percentage_change.

The whole job is one sorted(...) expression fed to csv.writer(...).writerows(...). A single shell pipe — os.popen("zcat "+d+"in/*.gz|grep ^[0-9]") — decompresses all 20 files as one stream and drops the comment and header lines (only data rows start with a digit), replacing the glob/gzip imports and the per-file loop. eval(a,{"NaN":0}) parses each flux array without json, and a single f>0 test keeps the valid values. Working in ratio space ((max/min-1) > 1 ⇔ > 100%) lets the ×100 happen once at write time, and comprehension for clauses stand in for statements throughout. A pure-standard-library variant (no zcat/grep) is 561 characters; the shell pipe takes it to 516.

Valid Flux Assumptions

The spec says to ignore “missing, null, NaN, or otherwise invalid” flux values. This solution treats a flux as valid only if it is present, numeric, finite, and strictly positive — the single f>0 test drops NaN (which eval maps to 0), zero, and negatives in one comparison, since NaN>0, 0>0, and any negative are all False. (For these 20 files the arrays in fact contain only NaN and positive floats — no nulls, zeros, or negatives — so f>0 is exactly sufficient.)

If a band has no valid fluxes, its min/max cells are left empty and only the other band contributes to percentage_change.

Verifiable Result

• 3 lines of code total — the ROUTINE header and one call in RunScript.mac, plus a single line in Flux.cls.
• Flux.cls is 516 characters (the Embedded Python method body is 463 characters; the rest is the class/method scaffolding), down from an initial 896.