mostlyright-data 0.15.0__tar.gz → 0.15.1__tar.gz

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mostlyright-data
-Version: 0.15.0
+Version: 0.15.1
 Summary: Mostly Right CLI for reproducible, reviewed Parquet datasets
 Project-URL: Homepage, https://mostlyright.md/
 Project-URL: Documentation, https://mostlyright.md/docs/guides/cli/

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "mostlyright-data"
-version = "0.15.0"
+version = "0.15.1"
 description = "Mostly Right CLI for reproducible, reviewed Parquet datasets"
 readme = "README.md"
 requires-python = ">=3.13,<3.14"

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/skills/mr-data-build/SKILL.md

@@ -102,7 +102,7 @@ absence as a defect.
 
 ## Commands
 
-These thirty are the whole surface, alongside `acquire-slices`, which names a capability
+These thirty-one are the whole surface, alongside `acquire-slices`, which names a capability
 gap rather than doing anything, and `export-hosted-candidate`, which is a backend operator step.
 
 | Command | What it does |
@@ -128,6 +128,7 @@ gap rather than doing anything, and `export-hosted-candidate`, which is a backen
 | `mr-data verify` | Hold one run's sealed table manifest against the parts it names, and report every disagreement. Without `--deep` it fetches the manifest alone; with `--deep` it fetches every part the run sealed and re-hashes it against the digest the manifest states. Exits non-zero when anything disagrees. |
 | `mr-data parts` | List one table version's parts — the identifier, the row count, the byte size and the recorded bounds of each — so a training loader can shard them across workers and read a table no single file should hold. `--version` names a version rather than the live one, and `--from`/`--to` keep only the parts a value range cannot rule out. Nothing is downloaded by listing them; see [training loaders](../../docs/TRAINING-LOADERS.md) for the pattern. |
 | `mr-data diff` | Compare two runs and say what changed. |
+| `mr-data connections` | List saved workspace connections and their current recipe coordinates; `--dataset ID` filters to connections granted to that dataset. No credential values are returned. |
 | `mr-data keys` | Enrol a source credential by name, list the names with their created and rotated times, and delete one. `keys set` reads the value from a file or from standard input and never from the command line; no command prints a value back. |
 | `mr-data stream` | Record a public `wss://` venue so a build can read it. `stream document register --file F` admits the connector document, `stream registry create --digest D` admits a set of them, `stream probe` listens briefly and seals nothing, `stream capture --seconds N` records a bounded window, `stream subscribe` keeps recording, `stream subscription {show|pause|resume|stop} ID` moves one, `stream status ID` reads a capture or a probe, and `stream recordings --document-digest D` names what a build could read and until when. `--seconds` is at most 600. `mr-data run` takes no duration: a build reads sealed batches and is bounded by them. |
 | `mr-data open` | Mint a single-use, short-lived address that continues this session in a browser at one dashboard path, and print it. `--open` hands it to the platform opener. The credential decides who mints: a device credential — what `mr-data login` stores — mints at Cloud, and a personal access token mints at Studio. |
@@ -1131,6 +1132,16 @@ Two things are currently broken here and are not worth diagnosing again:
 
 ## Source credentials
 
+**Check saved connections first when the user names a warehouse or API connection.** Run
+`mr-data connections --dataset DATASET_ID --json`. Use the matching active connection's ID and
+version digest in `source.connection`, with the query in `parameter_values` and a valid response
+reader; set `dataset.id` to the granted dataset. Do not ask the user to re-enter an already saved
+credential under Keys. If it is not granted, explain that an owner or administrator must allow
+the dataset in Settings → Connections & secrets. See [Saved connections](../../docs/RECIPE-CONNECTIONS.md)
+for the proposal shape. Studio resolves it into the canonical recipe and enforces access during
+execution. Named secrets below remain available for sources without saved connections.
+
+
 A source that needs a credential names it, and only the name ever appears in a recipe document, a
 receipt, an event or a log. Enrol the value once and reference the name:
 

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/acquisition/parsing.py

@@ -197,9 +197,10 @@ def parse_tabular_bytes_for_display(
     date, a time, a duration, a decimal, a not-a-number -- this one renders it into its exact text
     and carries on, so a column of dates can be *shown* rather than refused.
 
-    That is safe here and only here, because nothing this function returns is ever sealed. It has
-    exactly one caller, ``ux.peek``, which writes no bytes at all; ``tests/h3/test_parsing.py``
-    pins that there is no second one. Exactness is kept by
+    That is safe here and only here, because nothing this function returns is ever sealed. Its one
+    caller was ``ux.peek``, which wrote no bytes at all and which went with the retired local
+    execution lane, so it now has none; ``tests/h3/test_parsing.py`` pins that there is still no
+    caller, which is the same rule read at its stricter end. Exactness is kept by
     the rendering rather than by the refusal: the text is the value's own ``isoformat`` or ``str``,
     never a rounding, and the type of column a person is told about is unchanged -- a rendered
     ``date`` is still reported as a ``date``, because the text carries the name of the kind it came

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/cli.py

@@ -1792,6 +1792,9 @@ def _parser(
     from mostlyright.data_harness.thin import recipe as _recipe
 
     _recipe.declare_arguments(_add("recipe", help=_recipe.COMMAND_HELP))
+    from mostlyright.data_harness.thin import v4_connections as _connections
+
+    _connections.declare_arguments(_add("connections", help=_connections.COMMAND_HELP))
 
     from mostlyright.data_harness.thin import v4_dataset_covers as _v4_dataset_covers
 

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/sources/catalog/streaming_delta.py

@@ -1,10 +1,11 @@
 """Exact streaming delta classification over sorted authoring shards.
 
-The v1 classifier in :mod:`delta` reads one whole generation into dictionaries, and it explicitly
-requires the caller to hand it a complete provider-record-to-entry map.  At 548k provider records
+The v1 classifier this replaced read one whole generation into dictionaries, and it explicitly
+required the caller to hand it a complete provider-record-to-entry map.  At 548k provider records
 neither is possible: the map alone exceeds what ``canonical`` will serialize, and the entries alone
-exceed the plan's memory budget.  This module replaces that path for provider-scale sweeps while
-:mod:`delta` remains exactly as it is for bounded v1 fills.
+exceed the plan's memory budget.  This module took over that work for provider-scale sweeps, and
+the v1 classifier has since been retired with the bounded fill lane that was its only caller, so
+this is the whole of delta classification now.
 
 The shape of the work is a merge, not a join.
 

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/studio_boundary.py

@@ -85,7 +85,7 @@ PINNED_OPENAPI_SOURCE_SHA256 = "2b760333302cfddc52edb917eabd3d7d7859962d8053dd40
 #: One line, and it stays one line: ``_replace_exactly`` in the repin matches the whole assignment
 #: as a literal and refuses on any count but one, so a wrapped constant is a constant the repin can
 #: no longer rewrite.
-PINNED_V4_OPENAPI_SOURCE_SHA256 = "3e113fc44006370d8c6583bdea0564496a3d8bf591e8e4701c7de0d171eda015"
+PINNED_V4_OPENAPI_SOURCE_SHA256 = "ea530649071fb9b502ac0aa8a62b7bbea3ce0e39ab8078a68130644a72590f76"
 
 _DIGEST = re.compile(r"^sha256:[0-9a-f]{64}$")
 _IMAGE_DIGEST = re.compile(r"^[a-z0-9][a-z0-9._/-]{0,254}@sha256:[0-9a-f]{64}$")

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/thin/router.py

@@ -119,6 +119,7 @@ HOSTED_SUMMARY = dict(
             "create the page, correct it, write its record, watch it fill in, and retire it",
         ),
         ("recipe", "register one whole recipe and print the identifiers the server derived"),
+        ("connections", "list saved connections available for recipes"),
         ("cover", "generate and attach one branded cover image to a dataset"),
         ("run", "start one run against a registered recipe, under the ceilings you give it"),
         ("status", "report what one run is doing, what it delivered, and why it failed"),
@@ -257,6 +258,7 @@ def _hosted_parser() -> argparse.ArgumentParser:
     from mostlyright.data_harness.thin import recipe as recipe_module
     from mostlyright.data_harness.thin import (
         v4_artifacts,
+        v4_connections,
         v4_dataset_covers,
         v4_datasets,
         v4_narrative,
@@ -267,6 +269,7 @@ def _hosted_parser() -> argparse.ArgumentParser:
 
     v4_datasets.declare_arguments(_add("dataset"))
     recipe_module.declare_arguments(_add("recipe"))
+    v4_connections.declare_arguments(_add("connections"))
     v4_dataset_covers.declare_arguments(_add("cover"))
     v4_runs.declare_run_arguments(_add("run"))
     v4_runs.declare_status_arguments(_add("status"))
@@ -385,6 +388,7 @@ def _handlers() -> dict[str, Callable[[argparse.Namespace], dict[str, Any]]]:
         recipe,
         stream_venue,
         v4_artifacts,
+        v4_connections,
         v4_dataset_covers,
         v4_datasets,
         v4_handoff,
@@ -401,6 +405,7 @@ def _handlers() -> dict[str, Callable[[argparse.Namespace], dict[str, Any]]]:
         # module is plural because it holds the routes for all of them.
         "dataset": v4_datasets.dataset,
         "recipe": recipe.recipe,
+        "connections": v4_connections.connections,
         "cover": v4_dataset_covers.cover,
         **v4_tables.HANDLERS,
         "keys": v4_secrets.keys,

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/thin/v4.py

@@ -65,6 +65,7 @@ PINNED_V4_CONTRACT_VERSION = "4.1.0"
 
 #: ``POST`` -- one recipe document, registered in one transaction.
 CREATE_RECIPE_PATH = "/v4/recipes"
+LIST_CONNECTIONS_PATH = "/v4/connections"
 
 #: ``POST`` -- generate and attach one branded cover to an existing dataset.
 GENERATE_DATASET_COVER_PATH = "/v4/datasets/{dataset_id}:generate-cover"
@@ -412,6 +413,7 @@ DECLARED_V4_PATHS: frozenset[str] = frozenset(
         # closed -- so the recorded body lost it. That is what the cross-repo assertion is for.
         SECRET_PATH,
         LIST_SECRETS_PATH,
+        LIST_CONNECTIONS_PATH,
         # ⚠ AND THE LAST OF THEM, THE SESSION HANDOFF, promoted at the merge-time repin against
         # studio's main: studio's 404-02 shipped the mint while this milestone was in flight, so
         # `PENDING_V4_PATHS` is now empty and `mr-data open` calls a route that exists.
@@ -582,6 +584,12 @@ class StudioV4Client(StudioApiClient):
     # Recipes
     # ------------------------------------------------------------------------------------
 
+    def connections(self, *, dataset_id: str | None = None) -> dict[str, Any]:
+        query = {"workspace_id": str(self.session.workspace_id)}
+        if dataset_id is not None:
+            query["dataset_id"] = identifier(dataset_id, "dataset identifier")
+        return self._call("GET", LIST_CONNECTIONS_PATH, query=query, expected=(200,))
+
     def register_recipe(self, document: Mapping[str, Any]) -> dict[str, Any]:
         """Register one recipe document and return every identifier the call resolved.
 

mostlyright_data-0.15.1/src/mostlyright/data_harness/thin/v4_connections.py

@@ -0,0 +1,45 @@
+"""Read saved connection coordinates without returning credentials or modifying grants."""
+
+from __future__ import annotations
+
+import argparse
+from typing import Any
+
+from mostlyright.data_harness.thin.session import open_studio_session
+from mostlyright.data_harness.thin.transport import ThinLaneError
+from mostlyright.data_harness.thin.v4 import StudioV4Client
+from mostlyright.data_harness.ux.credentials import resolve_cloud_credentials
+
+COMMAND_HELP = "list saved connections for recipes, optionally filtered by dataset access"
+
+
+def declare_arguments(parser: argparse.ArgumentParser) -> None:
+    parser.add_argument("--dataset", help="only show connections granted to this dataset ID")
+
+
+def connections(
+    args: argparse.Namespace, *, client: StudioV4Client | None = None
+) -> dict[str, Any]:
+    selected = client or StudioV4Client(open_studio_session(resolve_cloud_credentials()))
+    response = selected.connections(dataset_id=args.dataset)
+    items = response.get("items")
+    if not isinstance(items, list):
+        raise ThinLaneError("THIN_RESPONSE_INVALID", "Studio returned no connection list")
+    fields = (
+        "connection_id",
+        "name",
+        "display_name",
+        "status",
+        "version_digest",
+        "origin",
+        "source_parameters",
+    )
+    return {
+        "schema_version": "mostlyright-thin-connections.v1",
+        "connections": [{key: item[key] for key in fields if key in item} for item in items],
+        "truncated": response.get("truncated", False),
+        "next_step": (
+            "Use connection_id and version_digest in source.connection. "
+            "The dataset must have an explicit grant; set dataset.id in the recipe."
+        ),
+    }

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/thin/vocabulary.py

@@ -193,6 +193,10 @@ CLASSIFICATION: dict[str, Lane] = {
         "it prints the record one run sealed about how it was built: every source it fetched and "
         "what those bytes hashed to, the coverage it delivered, and the raw snapshot's members",
     ),
+    "connections": Lane(
+        HOSTED,
+        "it lists saved connections and their recipe coordinates without exposing credentials",
+    ),
     "recipe": Lane(
         HOSTED,
         "it registers one whole recipe document -- the dataset, the question, the sources, the "

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/ux/commands/__init__.py

@@ -96,21 +96,11 @@ def _lazy(module_name: str, name: str, help_text: str) -> CommandSpec:
 
 
 _DISCOVERED: tuple[CommandSpec, ...] = (
-    _lazy(
-        "approve",
-        "approve",
-        "freeze one recipe by fingerprint, after showing you exactly what you are freezing",
-    ),
     _lazy(
         "auth",
         "auth",
         "inspect remotely validated device authority and token boundaries",
     ),
-    _lazy(
-        "author",
-        "author",
-        "turn a YAML recipe description into a recipe you can approve",
-    ),
     # The provider-scale workflow is three verbs -- harvest, then author, then publish -- and each
     # one is a separate mutation with a receipt of its own: `catalog-harvest` is the only one that
     # reaches a network or reads a key, and it is also the only one that can be asked for the key
@@ -121,11 +111,6 @@ _DISCOVERED: tuple[CommandSpec, ...] = (
         "catalog-author",
         "author one converged public catalog generation into bounded authoring shards",
     ),
-    _lazy(
-        "catalog_fill",
-        "catalog-fill",
-        "stage, reconcile, and publish one bounded public catalog successor",
-    ),
     _lazy(
         "catalog_harvest",
         "catalog-harvest",
@@ -136,98 +121,21 @@ _DISCOVERED: tuple[CommandSpec, ...] = (
         "catalog-publish",
         "classify one authored generation, publish it, and write its one composite receipt",
     ),
-    _lazy(
-        "catalog_search",
-        "catalog-search",
-        "search one exact verified sealed source catalog",
-    ),
-    _lazy(
-        "catalog_update",
-        "catalog-update",
-        "install one published catalog release after checking its exact fingerprint",
-    ),
     _lazy(
         "clarify",
         "clarify",
         "ask whether one clarifying question may be put to the user, and record the answer",
     ),
-    _lazy(
-        "deploy",
-        "deploy",
-        "stage one verified Build and activate it as the authenticated Editor",
-    ),
-    _lazy(
-        "deploy_dataset",
-        "deploy-dataset",
-        "bring the Build a deployment released back here and write its dataset notebook",
-    ),
-    _lazy(
-        "deploy_plan",
-        "deploy-plan",
-        "plan a live dataset request without deploying it",
-    ),
-    _lazy(
-        "deploy_status",
-        "deploy-status",
-        "report what Studio says about the run one deployment queued",
-    ),
-    _lazy("diff", "diff", "compare two builds and say what changed"),
-    _lazy(
-        "document",
-        "document",
-        "capture documents and request bounded auditable evidence through the local lane",
-    ),
-    _lazy(
-        "index",
-        "index",
-        "build one bounded private index from exact-verified local Builds",
-    ),
-    _lazy(
-        "inventory",
-        "inventory",
-        "list what this build can do: operations, types, joins, engines, and checks",
-    ),
-    _lazy(
-        "list_builds",
-        "list",
-        "list the builds under a folder, and say whether they were checked",
-    ),
     _lazy(
         "login",
         "login",
         "sign in to the Mostly Right cloud and store a device credential",
     ),
-    _lazy(
-        "peek",
-        "peek",
-        "open any file or public https address and show its columns, types, and first rows",
-    ),
-    _lazy(
-        "plan_check",
-        "plan-check",
-        "build this plan into a throwaway folder to see whether it works, then throw it away",
-    ),
-    _lazy("preflight", "preflight", "check what would stop a build before you start it"),
     _lazy(
         "record",
         "record",
         "seal already-recorded live messages into batches a build can read",
     ),
-    _lazy(
-        "review_setup",
-        "review-setup",
-        "create protected review material; this does not approve a Build",
-    ),
-    _lazy(
-        "search",
-        "search",
-        "search exact published public and private local stores without changing them",
-    ),
-    _lazy(
-        "show",
-        "show",
-        "show one build: what it answers, what a row is, its columns, checks, and fingerprints",
-    ),
     _lazy(
         "whoami",
         "whoami",

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/ux/headline.py

@@ -370,9 +370,6 @@ HEADLINES: dict[str, str] = {
     "CATALOG_TRACE_CANDIDATES": "The record of what this search looked at is not exact",
     "CATALOG_TRACE_PARTITION": "The groups this search sorted its results into are not the "
     "results it had",
-    "LOCAL_SEARCH_SERVICE_BINDING": "The Build this search chose is not the one that was "
-    "re-checked",
-    "LOCAL_SEARCH_TRACE_BINDING": "The record of this search does not match the Builds it read",
     "NEURAL_PIN_MISMATCH": "The search model on this machine is not the one it is pinned to",
     # Filling the Source Catalog and writing its entries. Both of these say "the payload", "the
     # facts and citation" or "the parser" in their own wording, which is a sentence written for

{mostlyright_data-0.15.0 → mostlyright_data-0.15.1}/src/mostlyright/data_harness/ux/readers.py

@@ -355,8 +355,10 @@ def require_a_build(run_dir: Path) -> None:
     still decides. Its only job is that pointing a command at the wrong folder reads like the typo
     it usually is.
 
-    Public because the comparison in :mod:`mostlyright.data_harness.ux.diffing` needs the same
-    signpost, and two copies of one sentence are two sentences that can drift apart.
+    Public because a second caller needed the same signpost, and two copies of one sentence are
+    two sentences that can drift apart. That caller was ``ux/diffing.py``, which went with the
+    retired local execution lane; the sentence stays stated once here rather than being folded
+    back into its one remaining reader, because that is the property worth keeping.
     """
 
     if not (run_dir / _RECEIPT).is_file():