ponder-lab / Hybridize-Functions-Refactoring / #3048
85%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 42
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 85.491%. Remained the same
#3048

push

github

web-flow 
Bump Ariadne to 0.52.95 (#941)

* Bump Ariadne to 0.52.94

Carries the numpy dtype resolution half and the fancy-index typing fix.

Three pins move back, and they move because the spellings resolve rather
than because a default returned.

`np.int` is what establishes that. Its run-time dtype is int64 and the
withdrawn default was float64, so int64 cannot be the old fallback
reappearing under another name. `np.long` reads int64 with it.
`np.float` reads float64 again, but float64 is what it meant and what
the default said, so on its own it distinguishes nothing; it is pinned
for coverage and the comment says so rather than letting it stand as
evidence.

Two spellings still emit nothing, both deliberately. `np.int16` has no
dtype enum value to map to, so no field reference can be written for it.
The builtin `bool` is outside the modeled family, where `np.bool` and
`np.bool_` sit inside it. Pinned as known absences: either one resolving
later is a result, not noise.

The route took two releases and neither half suffices alone. Withdrawing
the unconditional float64 default made the gap visible without closing
it, and a reader could not tell a resolved float64 from an unresolved
one before that. Defining the spellings closed it. The five pins here
record both steps, which is why the previous release moved the same five
in the opposite direction.

Suite 701/701.

* Advance the bump to 0.52.95

The release differs from 0.52.94 by one change, which models
`tf.keras.backend.floatx()` so a token spelled that way resolves instead
of degrading to an unknown dtype.

No pin moves, and that is the expected result rather than a null one.
Every dtype pin here spells its dtype as a numpy alias, so the suite
cannot observe a change confined to `floatx()` tokens. The suite passing
unchanged says the release did not disturb the numpy family, not that it
did nothing.

Where it should be observable is the whole-project population, on the
one subject of the six that reads `f... (continued)

3612 of 4225 relevant lines covered (85.49%)

0.85 hits per line

Jobs
ID Job ID Ran Files Coverage
1 #3048.1 42
85.49
Source Files on build #3048