Repository navigation
Figures, docstrings and the KB family set each subscript the way its standard prints it - #947
Conversation
github-actions
Bot
added
documentation
|
Important Review skippedWe couldn't safely recover the incremental review. No full review was started, and the last reviewed checkpoint was preserved. Retry later, or explicitly request a full review by commenting You can disable this status message by setting the Use the checkbox below for a quick retry:
Comment |
|
Sorry @jmrplens, your pull request is larger than the review limit of 150,000 diff characters |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## standards/iec-61094-5-shared-uncertainty #947 +/- ##
=========================================================================
Coverage 96.80% 96.80%
=========================================================================
Files 443 443
Lines 87888 87889 +1
=========================================================================
+ Hits 85081 85082 +1
Misses 2807 2807 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Numerical conformance
1893/1893 checks pass across 111 domains and 523 standards (227 normative designations, 125 further published sources). Used in the tables below is how much of that clause's published tolerance the deviation consumes: 100 % sits exactly on the limit, 5 % uses a twentieth of the allowance, and a dash means the clause states no two-sided tolerance for the quantity, so there is no budget to spend. It is reported and never used to decide a verdict, which is settled at full precision before any rounding. New checks (2)
Closest to their published limit (top 5) The rows with the least room left, so the ones a change is most likely to push over.
|
| Python Version | Tests | Failures | Coverage | Status |
|---|---|---|---|---|
| macos-latest-3.13 | 26898 | 0 | 96.8% |  Passed |
| macos-latest-3.14 | 26898 | 0 | 96.8% | Passed |
| ubuntu-latest-3.13 | 26898 | 0 | 96.8% | Passed |
| ubuntu-latest-3.14 | 26898 | 0 | 96.8% | Passed |
| windows-latest-3.13 | 26898 | 0 | 96.8% | Passed |
| windows-latest-3.14 | 26898 | 0 | 96.8% | Passed |
Full report at this commit: docs/CONFORMANCE.md · docs/conformance.json · full CI artifacts
…standard prints it `make subscripts` now reads the matplotlib figures each page embeds as well as the page and its diagrams, reads a run of letters as one symbol, and holds a family of letters that keep one meaning under several bases to one slope in a file. The places where a figure and its page disagreed are fixed on the side that lagged its printed standard; the KB family of DIN 45669-1, DIN 4150-2 and E DIN 4150-2, the r, u, o and e of the DIN 4150-2 assessment and the F and E of DIN 45672-2 set their descriptive parts upright across the docstrings, figures, diagrams, guides and errata register; the diagrams compose subscripts inside exponents; and the example fiches follow their standards. No public name, signature, default or returned value changes.
5a88c10 to
e88a846
Compare
|
1 participant
make subscriptsnow reads the matplotlib figures each page embeds as well as the page and its diagrams, from the mathtext source matplotlib keeps beside every label in the SVG, and it reads a run of letters as one symbol, soActivityAssessment.plot()draws, and the low-frequency differenceThe KB family has one set of subscripts across the library, the figures, the diagrams, the docstrings, the guides and the errata register:$KB_\mathrm{F}$ , $KB_\mathrm{Fmax}$ , $KB_\mathrm{FTm}$ , $KB_\mathrm{FTr}$ and $KB_\mathrm{FTm,Zug}$ with the descriptive parts upright, as DIN 45669-1, DIN 4150-2 and E DIN 4150-2 print them. The clock index of $KB_{\mathrm{FT}i}$ stays italic because the library sets an index that runs over a sum italic wherever it stands; the standards print it upright inside the subscript and italic under the sum.
DIN 4150-2:1999 and E DIN 4150-2 print the r of "Beurteilung" upright in the guide value$A_\mathrm{r}$ , the assessment time $T_\mathrm{r}$ and the clock count $N_\mathrm{r}$ , the same r as in $KB_\mathrm{FTr}$ , and the u, o and e of $A_\mathrm{u}$ , $A_\mathrm{o}$ and $T_\mathrm{e}$ upright as well. The docstrings, the figures, the diagrams, the guide snippets and the errata transcriptions now set all of them that way. $KB_\mathrm{FTr}$ beside an italic $A_r$ .
make subscriptsgains a table of letters that keep one meaning under several bases, and holds each family to one slope in a file, so a page can no longer setThe DIN 45672-2 quantities follow their print too: the running r.m.s.$\tilde v_\mathrm{F}$ and its maximum, the event value $v_\mathrm{E}$ and its hourly sum, the levels $L_{v\mathrm{F}}$ , $L_{v\mathrm{Fmax}}$ and $L_{v\mathrm{E}}$ , the band centre $f_{\mathrm{T}n}$ and the insertion loss $D_\mathrm{e}$ , in the docstrings, the figures and what $D_\mathrm{e}$ and $f_\mathrm{e}$ , and ges, Korr, Punkt and Zug.
TrainPassage.plot()andTrainPassage.plot_spectrum()draw. The docstrings of the E DIN 45672-3 prediction chain and of the E DIN 4150-2 train categories set upright the parts the two drafts print upright: E, BB, FB, DF, DB, the e ofThe diagrams can now set a subscript inside an exponent, so the energy sums they used to print as plain text are composed: the rating level of NT ACOU 112, the corrected levels of RD 1367/2007, Formula 5 of ISO 17497-2, Formula 26 of EN 12354-1, Formula 12 of EN 15657, Formula 17 of EN 12354-5, Formulae A.5 to A.7 of the DB-HR and the calibration factor. The exponential detector and the two-ray sum are written with exp(), the differential d of the waveguide bottom condition is upright as ISO 80000-2 asks, the test voltage$U_\mathrm{p}$ of IEC 60268-5 has its upright p, and the letters of an exponent stay italic. The theory page that embeds the flanking diagram writes $R'_\mathrm{w}$ with the upright w EN 12354-1 prints.
The example fiches follow their standards: the ISO 3382 fiche heads its centre time column$T_\mathrm{S}$ , the ISO 9614 and IEC 61043 fiches set the p and I of $F_{pI}$ , $F_{pI_\mathrm{n}}$ , $F_{p|I_\mathrm{n}|}$ and $\delta_{pI0}$ italic as ISO 9614-2 and ISO 9614-3 print them, the ISO 15186 fiches set the I of $R_{I,\mathrm{w}}$ and $D_{I,\mathrm{n,e}}$ italic in the table, the plot axis and the note on element units, and the RD 1367 fiche shows the upright Keq and K of its plot legend.
The smaller items: the diffusers and metadiffusers guides write the normalised coefficient$d_\mathrm{n}$ of ISO 17497-2, 3.12, where they wrote $d_\mathrm{norm}$ , beside the italic well depth $d_n$ , so they now agree with the goniometer diagram, and the metadiffuser diagram writes it the same way; the areas, distances and velocities of ISO 11820 are upright in $v_0$ with a zero in the docstrings, the resilient-layers guide and its figure, where it was $v_o$ or $v_\mathrm{h}$ ; and the centre time of ISO 3382-1 is $T_\mathrm{S}$ , with the capital S of its Table A.1, in the guides, the glossary, the figure and the fiche, apart from the structural reverberation time $T_\mathrm{s}$ .
silencer_in_situ; the hammer velocity of Hopkins's tapping model isNothing breaks. No public name, signature, default or returned value changes; what changes is how symbols are typeset in docstrings, guides, figures, diagrams, the errata register and the example fiches. For contributors,
make subscriptsis stricter: a figure label that sets a subscript the other way from the page that embeds it fails the check, and so does a member of a linked family that leans while its siblings stand. Regenerate the figure or the diagram after changing a label, and name the symbol in theuprightkey of a diagram label where the composer would otherwise keep a single letter italic.Two choices are left open on purpose. The KB base itself has two slopes because its sources disagree: DIN 45669-1:2010-09 prints KB upright, while DIN 4150-2, E DIN 4150-2, DIN 45672-2 and E DIN 45672-3 print it italic; the diagrams set it upright and the figures and the errata register italic. EN 15657 prints the v of$L_\mathrm{v}$ upright, and the guide, the module and the diagram keep the italic $L_v$ the rest of the corpus uses for a velocity level; changing it means taking the whole EN 15657 symbol set at once.
Verified with the full test suite on Python 3.13 (26827 passed, 69 skipped), every documentation code block (7483 blocks on 730 pages), the site type check, build and HTML validation, ruff, ruff format, mypy on src and scripts, bandit, and the repository checks:
make subscriptspasses on all 1282 files and the images they embed and fails with the 21 disagreements on the tree this branch starts from; all 3312 committed figures and all 77 example fiches match their generators; the figure checks pass on all 1622 drawings; all 6397 mathtext labels parse; the conformance report regenerates with no change; and the API reference, llms text and errata pages are regenerated. Each of the 30 changed figures and diagrams (18 diagrams and 12 figures, four variants each) and each of the 7 changed fiches was looked at in both languages.