Check source selection and complex-port order before a handoff
Trace a routing decision to its final source rule and a measured Sij value to its original token pair before sending reviewers or plotting data.
A final owner line is a source decision
A PR grouping workflow can route many paths to the same declared owner, but selecting the wrong branch snapshot invalidates that grouping before matching begins. Record the chosen source name and SHA-256, the changed-path list hash and the final rule line. The owner-group list indexes every original path occurrence, so duplicate paths do not disappear.
An empty owner list is a decision with provenance. If /apps/ names @apps and a later /apps/github has no owners, the final status is explicitly-unowned. A path with no matching valid line has unmatched status. Keep both categories visible instead of turning them into a generic missing-owner label.
A CSV column can look correct while holding the wrong Sij
A two-port vector has four pairs, but its legacy order is column-major: S11,S21,S12,S22. Label each row and column before extracting S21 or plotting a curve. For three and four ports, verify physical matrix row grouping and the first-row frequency token.
For the starter first point, S21 is raw0.8 and -0.2, not the following0.02 and0.01 pair. Frequency1 is in MHz and therefore1,000,000Hz. The chosen R75 remains a declared reference; the tool does not normalize to50Ω or calibrate a VNA. Raw tokens and byte locations let a receiver check these decisions without relying on rounded display values.
Inspect a gap before automating the next step
Skipped CODEOWNERS invalid lines appear in the complete report. Unsupported legal escaped patterns abort routing. Unsupported Touchstone features abort matrix extraction. Neither result should be converted into a claim that GitHub requested reviewers or that a circuit passed RF analysis.
Copying a large report gives only the labelled preview. Send the complete downloadable JSON/CSV together with the original source and the supported-profile note. All reported pairs and paths are local declarations or measurements, not executed scripts.
| Handoff | Check | Remaining gap |
|---|---|---|
| Declared owners | Final pattern/line, all owners, explicit clearing | GitHub base branch, access and account existence |
| Measured S matrix | Port coordinates, units, R, raw tokens and source offsets | RF calibration, circuit/device validity |
| Large report | Full native files and original source hashes | Copy remains a200-row preview |
- Choose the snapshot and port count explicitly.
- Verify one nontrivial path and one off-diagonal S parameter against the original source.
- Keep the complete outputs when handing work to another person.
References
- CODEOWNERS changed-file grouping request
First-person demand read2026-10-07; offline access verification remains outside scope.
- S-parameter CSV plotting question
First-person demand read2026-10-07; matrix extraction supports the stated handoff, without circuit validation.
- GitHub CODEOWNERS syntax
Correctness reference read2026-10-07, not demand evidence.
- IBIS Touchstone2.1 specification
Includes1.0/1.1 compatibility; read2026-10-07. This tool declares a narrower physical-row S profile.
Tools in this category
Expand a tool to see its steps, options and supported formats, then open its workspace.
CODEOWNERS changed-path routingResolve each supplied changed path against one selected CODEOWNERS snapshot, retaining the last valid rule, all declared owners and explicit ownership gaps.
Group changed paths by declared reviewer without accessing GitHub. Choose the intended branch snapshot yourself; inspect skipped source lines and distinguish cleared owners from unmatched paths.
Steps
- Choose the intended CODEOWNERS snapshot, then enter repository-relative changed paths, one per line.
- Run routing; inspect skipped rule count, declared owners, explicitly unowned paths and unmatched paths.
- Download full JSON/CSV with source provenance and all owner groups before using the handoff.
Available options
- Protect CSV formula-like text
- On by default
Capabilities and limits
- One explicitly selected UTF-8 CODEOWNERS file or paste up to 2 MiB, plus a newline-delimited changed-path list up to 2 MiB. Choose the intended snapshot; .github/root/docs priority and pull-request base branch are not inferred.
- Up to 10,000 rule lines, 10,000 paths, 1,024 UTF-16 units per pattern/path and 100 owners per rule. Valid rule count × supplied path count must be at most 2,000,000. Complete JSON and CSV together must fit 32 MiB; combined limits apply.
- Case-sensitive GitHub-style *, ?, whole-component **, root paths and directory patterns. Last matching valid source line wins. Owners on the same line stay together; an empty owner list explicitly clears ownership. Duplicate paths remain in order. User, team and email strings are declarations; account existence, access and review requests remain unverified.
- GitHub-invalid negation, character ranges and escaped leading # lines are skipped with full original lines, reasons and UTF-8 byte offsets. Legal escaped patterns, including escaped spaces, fail the whole batch as Unsupported. Absolute/backslash/dot-dot/control/ambiguous empty path segments reject; source and path text remain inert.
- The complete report includes original sources and hashes, valid/invalid rules, every supplied path and owner grouping. Table previews 200 rows, clipping long cells to 2,000 UTF-16 units. Above 20,000 report characters, copying gives a labelled preview; full downloads retain all rows. Cancellation or an error publishes no partial routing.
Touchstone S-parameter matrix reportRead one declared 1–4-port Touchstone S-parameter file into a complete matrix CSV/JSON, preserving raw numeric tokens, option provenance and source positions.
Export every Sij measurement in the correct port order before plotting. Inspect complex values, magnitude, phase and dB alongside the original RI, MA or DB tokens; no device or circuit is simulated.
Steps
- Select the ASCII .sNp source and explicit port count; confirm the option line and physical matrix rows.
- Inspect frequency, reference, representation and matrix coordinate summaries.
- Download all complex pairs as JSON/CSV, retaining raw measurements before plotting or further analysis.
Available options
- Explicit port count
- 1 port · 2 ports · 3 ports · 4 ports
- Protect CSV formula-like text
- On by default
Capabilities and limits
- One ASCII file or paste up to 10 MiB with explicit 1–4 ports. A recognized .sNp filename must match the selected port count. Up to 20,000 strictly increasing frequency points, 100,000 complex pairs and 4,096 UTF-16 units per numeric token; complete JSON/CSV together up to 40 MiB. Combined limits apply.
- Visible Version1.0/1.1 compatible S-parameter profile: exactly one pre-data # option line. An empty # means GHz, S, MA, R50. A positive finite scalar R may occur in any supported option order. Missing # is invalid; second/late options, per-port R, Y/Z/G/H, noise records and 2.x keywords are Unsupported profile.
- Physical rows: 1/2-port points keep frequency and every complex pair on one line; 3/4-port points use one matrix row per physical line, frequency only in the first row. Other valid wrapping is Unsupported rather than labelled a standard error. LF/CRLF, inline ! comments and original ASCII source are preserved; standalone CR and non-ASCII input are unsupported.
- Two-port source order is S11,S21,S12,S22; 3/4-port order is matrix-row order. Full raw token strings and line/column/byte coordinates accompany derived float64 real, imaginary, magnitude, normalized phase degrees and 20log10 magnitude dB. Exact zero has phase0 and explicit string -Infinity dB. Nonfinite input, float64 overflow/underflow and unsupported ranges reject rather than fabricate values.
- Complete JSON preserves option defaults, reference, source hash, comments and every matrix pair. CSV retains all frequency/pair rows and safe spreadsheet text. Table previews 200 pairs, clipping cells to 2,000 UTF-16 units; above 20,000 report characters, copy is a labelled preview. No RF calibration, circuit performance, reciprocity or device validation.