build(deps): bump log from 0.4.22 to 0.4.25#2439
Merged
Merged
Conversation
Bumps [log](https://github.com/rust-lang/log) from 0.4.22 to 0.4.25. - [Release notes](https://github.com/rust-lang/log/releases) - [Changelog](https://github.com/rust-lang/log/blob/master/CHANGELOG.md) - [Commits](rust-lang/log@0.4.22...0.4.25) --- updated-dependencies: - dependency-name: log dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>
dependabot
Bot
added
dependencies
10 tasks
jqnatividad
added a commit
that referenced
this pull request
May 25, 2026
…nputs (#3901) * refactor(profile): port DP+ jinja2 helpers from Python to Rust (minijinja) Replaces the PyO3-based formula engine with a native Rust implementation built on minijinja. The Python files are not yet deleted (Phase 0d) and the python feature gate is still in place (Phase 0e), but the runtime code path no longer touches Python — qsv profile evaluates spec formulas entirely in-process. * formula_engine.rs preserves the evaluate_spec / FormulaResult API exactly so profile.rs::merge_formula_results was untouched beyond the module import. * formula_helpers.rs ports all 17 DP+ helpers; the 2 SQL-backed ones (temporal_resolution, guess_accrual_periodicity) return errors today, matching the old Python-stub behavior; Phase 0c wires them to Polars. Verified: 22 new unit tests pass, all 5 existing integration tests pass (including the spatial_extent_wkt end-to-end formula round-trip). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * refactor(profile): remove Python; wire SQL helpers to Polars (Phase 0c-e) Phase 0c: temporal_resolution and guess_accrual_periodicity now query the input CSV via Polars SQL (new src/cmd/profile/sql_backend.rs). Each helper call rebuilds a fresh LazyFrame + SQLContext, registers the CSV as table `data`, runs a CAST-to-VARCHAR DISTINCT/ORDER BY query, then parses the result strings as dates and computes interval thresholds identical to DP+'s Python semantics. The backend is installed on the current thread by evaluate_spec for the render-pass duration and cleared after; thread_local storage avoids cross-test contamination. Phase 0d: deleted src/cmd/profile/py/ and src/cmd/profile/py_engine.rs; removed the jinja2_helpers.py entry from .github/workflows/devskim.yml. Phase 0e (partial): profile feature is now [feature_capable, polars, yaml_serde] — python is out, polars is in. profile is removed from the python pull in distrib_features. Adding profile to datapusher_plus is DEFERRED: it triggers feature-cfg overlap in sortcheck.rs (and likely other files) since the profile feature carries feature_capable. Needs a broader feature-cfg audit; queued as a follow-up. USAGE banner updated to drop the python3+jinja2 requirement language. docs/help/profile.md regenerated. Two tangential clippy-fix nits in describegpt.rs and moarstats.rs picked up by --fix. Verified: 29 unit tests pass (including 6 new SQL-backed tests), all 5 integration tests pass, cargo +nightly fmt + cargo clippy clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * refactor(profile/dcat): split build() into per-section helpers (Phase 1) Pure refactor — bit-identical output. The build() function becomes a thin orchestrator over six dedicated helpers: add_context_and_type — @context + @type add_core_identity — dct:title/description/identifier/modified/issued add_provenance — dct:license + dct:publisher add_classification — dcat:keyword + dcat:theme add_coverage — dct:spatial + dct:temporal add_governance — dcat-us:accessLevel add_distributions — dcat:distribution array Each helper has an inline NOTE comment pointing at the upcoming Phase 2 shape change (spatial/temporal → array, license → Distribution, conformsTo as Standard object, language ISO 639-1) so subsequent diffs land in the obvious spot. Adds a take_first_str(obj, &[primary, fallback]) helper to replace the string_opt(...).or_else(...) chains scattered through the dataset builder. Verified: 4 dcat unit tests + 5 integration tests pass; spatial/temporal JSON shapes and key-order are preserved exactly so downstream consumers see no change. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile/dcat): DCAT-US v3 shape migrations (Phase 2) Breaking changes to the dcat output shape per the DCAT-US v1.1 → v3 migration guide (https://resources.data.gov/resources/dcat-us-3-migration/). * 2a: dct:spatial is now an array of dct:Location objects (was a single object). Affects both the WKT-suggestion branch and the bbox fallback. * 2b: dct:temporal is now an array of dct:PeriodOfTime, with one entry per inferred date column (previously only the first DATE_FIELDS entry was consumed). * 2c: dct:license moved from Dataset to Distribution. Read order: resource.license_id → resource.license → package.license_id → package.license. New CLI flag --dcat-legacy-license re-emits at the Dataset level for transitional back-compat (default off). * 2d: dct:conformsTo is always emitted as a dct:Standard object pointing at https://resources.data.gov/resources/dcat-us3/. dct:language, when provided, is normalized to ISO 639-1 (en-US → en, "English" → en); unrecognized values are dropped (Phase 5 will warn instead). * 2e: dct:modified rejects ISO 8601 interval syntax (R/P1Y, P1Y, start/end ranges) — DCAT-US v3 requires a discrete date here. Frequency-of-update values belong on dct:accrualPeriodicity (Phase 5). The two existing integration-test assertions that touched dct:spatial were updated to index into the array. Added 7 new dcat unit tests covering each shape change. Verified: 11 dcat unit tests pass (4 existing + 7 new), all 5 integration tests pass, format + clippy clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile): accept http(s) URL inputs, download to tempfile (Phase 3a) When the input arg looks like a URL (http:// or https:// prefix, case-insensitive), download it to a NamedTempFile via reqwest::blocking and feed the local tempfile path to the rest of the pipeline (stats, frequency, sqlp-backed helpers). The tempfile handle is held in a local binding so its Drop fires only when run() returns. The original URL is preserved and stamped onto resource.url (if the resource doesn't already declare one) so the DCAT projection's dcat:downloadURL slot gets populated automatically. The existing absolute-IRI guard in build_distribution keeps non-IRI inputs from polluting the slot. DCAT-markup discovery (Link: rel=describedBy, sibling JSON-LD URLs, HTML script-tag JSON-LD) is queued for Phase 3b. The --no-dcat-discovery and --dcat-discovery-timeout flags also land with 3b. Verified: new url_detection unit test passes; all 36 existing profile unit tests + 5 integration tests still pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile): DCAT-markup discovery on URL inputs (Phase 3b) New module src/cmd/profile/dcat_discover.rs implements best-effort DCAT-US v3 sniff for URL inputs. Ships with the most authoritative mechanism — HTTP Link: rel=describedBy header — and is structured so the two remaining mechanisms (sibling URLs by convention, JSON-LD <script> blocks in HTML landing pages) can land as follow-up commits without API churn. * RFC 8288 Link header parser handles multi-token rel values, multiple comma-separated links, and case-insensitive rel matching. * Relative IRIs resolve against the original URL via the url crate. * extract_dcat_dataset handles the three common shapes: bare object, @graph array, and shape-fallback for non-conforming publishers. * All network / parse errors are non-fatal — discovery is enrichment; failures fall through silently. * Discovered DCAT surfaces under output.dcat_discovered for now. Phase 4 will merge it with the auto-inferred projection per the documented precedence chain (force:true > discovered > inferred > plain seed > formulas). New CLI flags wired through Args: --no-dcat-discovery --dcat-discovery-timeout <secs> (default 5) 12 new unit tests cover Link-header parsing, IRI resolution, and the dataset-extraction shapes. All 49 profile unit tests + 5 integration tests pass; format + clippy clean. docs/help/profile.md regenerated. One nightly-fmt drift in describegpt.rs picked up too. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile): --initial-context replaces --package-meta/--resource-meta (Phase 4a) Single unified JSON input replaces the two old seed-meta flags. Top-level keys: * package — CKAN-shaped seed for the dataset block * resource — CKAN-shaped seed for the resource block * dataset_info — RFC 6901 JSON-Pointer overrides into the final output dataset_info is the escape hatch: each entry sets a value at the named pointer in the assembled output JSON, applied last so it wins unconditionally over inference, discovered DCAT, the CKAN block, and formula output. Missing parent objects are auto-created; non-pointer keys and malformed paths are silently skipped. The fixture tests/resources/profile/dcat-init-context.json documents every field the projection currently reads or will read in Phase 5, with a sibling README mapping each slot to its DCAT-US v3 target property + a per-property `force` semantics primer for Phase 4b. Per-property `{value, force}` wrapper detection and the merge with discovered DCAT (the actual precedence chain — force:true > discovered > inferred > plain seed > formulas) land in Phase 4b. Verified: 4 new unit tests for apply_pointer_overrides + 1 new integration test (profile_initial_context_seeds_package_and_overrides_via_dataset_info) exercise the full --initial-context flow end-to-end including JSON-Pointer overrides, ISO 639-1 language normalization (en-US → en), ISO 8601 interval rejection on dct:modified, and license-on-Distribution. All 49 unit + 6 integration tests pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile): wrapper detection + discovered-DCAT merge (Phase 4b) * load_initial_context now normalizes per-property {value, force} wrappers. The wrapper is detected by exact two-key shape (value + boolean force), so structural fields like contact_point: {fn, hasEmail} pass through untouched. Nested wrappers (inside Maps or Arrays) unwrap recursively. * merge_discovered overlays publisher-stated DCAT onto the inferred projection with "fill gaps" semantics: inferred values (including --initial-context seed values) always win on collision; discovered fills only the slots inferred left absent. @context, @type, and dcat:distribution are never overwritten (per-distribution merging needs an identity scheme first; out of scope). * The raw discovered DCAT still surfaces under output.dcat_discovered for diffing / auditing. The force flag is currently accepted forward-compat but no-op under fill-gaps — since seeded values pre-populate inferred, plain values and force:true wrappers produce identical output today. Full override-discovered semantics can layer onto merge_discovered without an init-context schema break. Verified: 7 new wrapper-normalization tests + 3 new merge tests; all 63 profile unit tests and 6 integration tests pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile/dcat): add all DCAT-US v3 mandatory + recommended fields (Phase 5) dcat::build signature changes from `-> Value` to `-> (Value, Vec<DcatWarning>)`. profile.rs::run threads the warnings into the top-level output JSON as `dcat_warnings` (elided when empty). Each entry is {field, severity, message} where severity is Required (mandatory v3) or Recommended. Dataset-level additions: * dcat:contactPoint (vcard:Individual) — MANDATORY. Reads package.contact_point.{fn, hasEmail}; falls back to package.maintainer + package.maintainer_email. Missing → Required warning. * dcat-us:bureauCode / programCode — accepts arrays or comma-separated strings. Missing → Recommended warning each. * dct:accrualPeriodicity — slug → EU controlled-vocab IRI (annual, monthly, daily, etc., plus R/P* aliases). Also reads dpp_suggestions.accrual_periodicity for formula-derived values. * dcat:temporalResolution — pass-through ISO 8601 duration; reads dpp_suggestions.temporal_resolution for formula-derived values. * dct:accessRights, dct:rights, dcat:landingPage (IRI-validated), dcat:describedBy (IRI-validated), dcat-us:purpose, skos:scopeNote, dcat-us:liabilityStatement, dcat:inSeries (IRI-validated). Distribution-level additions: * dcat:accessURL (IRI-validated), dct:rights, dct:modified, dcat-us:accessRestriction / useRestriction / cuiRestriction (structured objects, passed through verbatim). Tests: 9 new dcat unit tests (contact-point happy-path/fallback/missing, US-codes array/csv/missing, accrual-periodicity slug mapping, extended metadata pass-through, distribution v3 additions) + 2 new integration tests (full-v3 population emits every mandatory+recommended slot with no warnings; missing contactPoint surfaces a Required-severity warning). All 20 dcat + 8 integration tests pass; format + clippy clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(profile): opt-in JSON Schema validation (Phase 6) New src/cmd/profile/dcat_validate.rs exposes validate_dataset(value) backed by jsonschema = 0.46 (already a workspace dep). Returns one DcatWarning per violation — Required severity when the schema reports a `required` violation, else Recommended. The schema is an embedded minimal v3 enforcement of the mandatory keys per https://resources.data.gov/resources/dcat-us3/: @type=dcat:Dataset, dct:title/description/identifier (non-empty), dct:publisher (with foaf:name), dcat:contactPoint (vcard:Individual with vcard:fn + mailto-prefixed vcard:hasEmail), dct:conformsTo (dct:Standard with @id), and dcat:distribution (≥1). Recommended fields are intentionally NOT enforced by schema — those are surfaced by the in-projection helpers (add_contact_point, add_us_codes, etc.) which give richer guidance per missing field. Follow-up: vendor the full GSA dcat-us jsonschema bundle from https://github.com/GSA/dcat-us/tree/main/jsonschema under resources/dcat-us-v3/ pinned to an upstream commit SHA and switch embedded_minimal_schema() to load it via $ref resolution. The Validator::options().build pattern already mirrors src/cmd/validate.rs for an easy swap. CLI flags wired through Args: --validate-dcat Append schema violations to dcat_warnings. --strict-dcat Under --validate-dcat, fail the command instead. 5 unit tests + 3 integration tests cover: passing minimal dataset, missing contactPoint surfacing Required severity, bad email format (mailto check), wrong @type rejection, missing distribution array, full-init-context validates clean, missing-cp triggers warning, and --strict-dcat fails the command without writing the output file. docs/help/profile.md regenerated. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(profile): address roborev #2439 — 6 Medium findings 1. SqlBackend now honors --delimiter and --no-headers. Added with_delimiter / with_has_header builders and threaded ctx_args values through evaluate_spec. SQL-backed helpers (temporal_resolution, guess_accrual_periodicity) now see the same columns stats/frequency saw on TSV/SSV/no-header inputs. 2. URL downloads stream via std::io::copy instead of buffering the full response with response.bytes() — large remote CSVs no longer risk OOM. 3. New tempfile_suffix_for_url helper parses the URL via url::Url, uses Url::path() to strip query strings and fragments, and preserves CSV-family compound extensions (.csv.gz, .tsv.gz, .csv.zst, .csv.bz2, .csv.xz, …). Previous Path::extension() on the raw URL turned `data.csv.gz?token=x` into garbage and dropped the .csv from compound extensions. 4. Reordered profile.rs::run so --initial-context dataset_info JSON-Pointer overrides apply BEFORE --validate-dcat / --strict-dcat runs. A user supplying a missing mandatory field via /dcat/dcat:contactPoint now satisfies validation. Build-time warnings are stashed under __pending_dcat_warnings during the intermediate phase and unstashed after the schema pass. 5. set_by_pointer now descends into existing arrays via numeric path segments. Previously /dcat/dcat:distribution/0/dct:license replaced the distribution array with {"0": {...}}, corrupting the DCAT shape. Out-of-range indices and non-numeric tokens against an array silently skip rather than convert it to an object. 6. Added Kwargs support to truncate_with_ellipsis, format_number, format_date, format_range, format_coordinates, and spatial_extent_feature_collection. Existing DP+ formulas using `format_date(format='%B %d, %Y')`, `truncate_with_ellipsis(length=5, ellipsis='…')`, `spatial_extent_feature_collection(name=, bbox=, feature_type=)` etc. now work. spatial_extent_feature_collection uses a kwargs-only signature (minijinja can't cleanly bind typed positional + Kwargs in an all-kwargs call); the docstring documents the constraint. Tests: 4 new sql_backend (TSV, no-header roundtrip), 5 new profile.rs (URL suffix compound/query/array-pointer regressions), 1 kwargs end-to-end, 1 dataset_info-rescues-strict integration. All 86 unit + 12 integration tests pass. DcatWarning + Severity gained #[derive(Deserialize)] so the intermediate stash/unstash round-trips cleanly. Closes roborev job 2439. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(profile): derive dct:title from URL basename, not tempfile stem When the input is a URL, the title slots (dct:title on Dataset, dct:title on Distribution) previously surfaced the random tempfile suffix (qsv-profile-XkZGBK) since context::build seeds resource.name from the local file path and dcat::build's title fallback uses Path::file_stem on the same. Both are useless to downstream consumers. New url_title_default(url) helper parses the URL with url::Url, strips CSV-family compound extensions (.csv.gz, .tsv.gz, .csv.zst, …) and single extensions, then returns the last non-empty path segment. Opaque/UUID basenames (e.g. CKAN's /datastore/dump/<uuid>) pass through unchanged — still better than the random tempfile suffix and traceable back to the input. Host-only / malformed URLs return None and the caller falls back to today's tempfile-stem default. profile.rs::run seeds: * package.title — via .entry().or_insert(), so a real seed wins. * resource.name — replaced only when its current value matches the tempfile stem context::build would have inserted; user-supplied values via --initial-context or formulas survive untouched. Verified live against the WPRDC Pittsburgh 311 endpoint (https://data.wprdc.org/datastore/dump/5202679a-...): both dct:title and the Distribution dct:title now read "5202679a-d243-402e-b82a-63189995a942" (the UUID basename) instead of "qsv-profile-PL3u1k" (the tempfile stem). qsv:sourcePath still records the tempfile path for traceability. Tests: 6 new url_title_default unit tests + the 5 existing URL-suffix ones still pass. All 92 profile unit + 12 integration tests pass. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(profile): address roborev #2440 — 3 Medium findings 1. profile.rs: stashed build-time dcat_warnings are now filtered against the final dcat block before being re-emitted. New helper final_dcat_has_field looks up each warning's field name (top-level key, or JSON-Pointer path for nested fields) in the post-override dcat snapshot and drops the warning if the slot is populated. Result: a `dataset_info` override that supplies a previously- missing mandatory field no longer leaves a "missing X" warning stale in the output. 2. context.rs::load_initial_context now applies normalize_value_force to dataset_info too — previously only package and resource were unwrapped. A documented override like "/dcat/dcat:contactPoint": {"value": {...}, "force": true} now unwraps to the inner Value before being written to the output via set_by_pointer, so the wrapper itself doesn't become the dcat:contactPoint value. This rescues --strict-dcat instead of tripping it. 3. spatial_extent_feature_collection regained positional-arg support alongside kwargs. minijinja's Function impl doesn't route `Rest<Value> + Kwargs` cleanly — Rest greedily consumes the kwargs container as its only positional, leaving Kwargs empty (confirmed via debug print: `args.len()=1, kwargs.args()=[]`). Workaround: accept Rest<Value> only and detect a trailing kwargs-shaped Value ourselves via the public Kwargs::try_from impl, splitting it off as kwargs and processing the leading slice as positionals. Both call styles documented in DP+'s docstring now work: spatial_extent_feature_collection("Name", bbox, "manual") spatial_extent_feature_collection(name="X", bbox=[...], feature_type="m") spatial_extent_feature_collection("Mix", bbox=[...], feature_type="m") Tests: 2 new helper unit tests (positional + mixed-pos-and-kw call styles) and 2 new integration tests (stale warnings cleared by dataset_info override; wrapped {value, force} dataset_info override rescues --strict-dcat). All 94 profile unit tests + 14 integration tests pass; format + clippy clean. Closes roborev job 2440. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * typos: add more exceptions * docs(features): drop `python` from all_features + distrib_features CI docs-drift-check caught a docs gap from the Phase 0e Cargo.toml change: `profile` no longer depends on `python`, and `python` was dropped from the `distrib_features` enumeration in Cargo.toml, but the two doc lines in docs/FEATURES.md still listed it. `python` is still a defined feature (`python = ["pyo3"]` exists for the standalone `py` command), so this is purely a docs-enumeration sync — no Cargo.toml or behavior change. Verified: scripts/docs-drift-check.py reports "no drift detected". Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(features): include `python` in distrib_features exception clause Roborev 2444#1: the prior commit dropped `python` from the explicit distrib_features enumeration but left the surrounding prose claiming distrib_features is "all features except self_update, ui and magika". Since `python` is also excluded now, add it to the exception clause so the description matches reality. Verified: scripts/docs-drift-check.py reports "no drift detected". Closes roborev job 2444. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * address review: Copilot PR #3901 + devskim suppressions Six Copilot review findings: * formula_helpers.rs module docs: drop the stale "SQL globals currently short-circuit to an error" claim — Phase 0c shipped the Polars SQL backend and the helpers are wired through it. * mode() determinism: previously HashMap-based, so ties were broken non-deterministically and could change guess_accrual_periodicity output across runs. Now tracks first-seen index alongside the count and breaks ties via Reverse(first_seen) so the earliest-encountered value wins (matches Python Counter.most_common(1)). * parse_date_strings now accepts ISO 8601 datetimes with fractional seconds — %Y-%m-%dT%H:%M:%S%.f and %Y-%m-%d %H:%M:%S%.f — before falling through to second-precision / date-only / RFC 3339. DP+'s datetime.fromisoformat accepts these and temporal_resolution / guess_accrual_periodicity were hard-failing without them. * resolve_input now uses util::create_reqwest_blocking_client (same helper validate / describegpt / fetch use) for consistent user-agent, gzip/brotli/zstd compression, rustls, and 503 retry. * dcat_discover::discover swaps to util::create_reqwest_blocking_client for the same reasons. The describedBy fetch in discover_via_link_header is now capped at 4 MiB via std::io::Read::take, so a publisher-controlled Link target can't blow up qsv's memory if it points at a multi-GB resource. * dcat-init-context.README.md status column: every "⏳ Phase 5" entry flipped to "✅ today" since the Phase 5 fields all landed; the surrounding prose updated to match. GitHub Advanced Security devskim noise (17 findings): * The http:// IRIs in accrual_periodicity_iri (EU publications.europa.eu frequency vocab) are stable opaque identifiers published with the http scheme by spec, same as the Creative Commons / Open Data Commons IRIs in license_iri. Added `// DevSkim: ignore DS137138` to each. * Same situation for the W3C DCAT canonical type IRI `http://www.w3.org/ns/dcat#Dataset` in dcat_discover::is_dcat_dataset and its two extract_dataset_* test fixtures. * The two bare http:// strings in url_detection_recognizes_http_https_case_insensitive are detector inputs — exactly what is_http_url is testing — so the test gets DS137138 suppressions too. * The two "TODO Phase 3b'/3b''" markers in dcat_discover.rs's module docstring tripped devskim's "suspicious comment" rule (DS176209); reworded to "(Phase 3b' follow-up)" / "(Phase 3b'' follow-up)" — same intent, no TODO/FIXME keyword. Tests: 2 new mode/parse-date unit tests + all 96 unit + 14 integration tests pass; format + clippy clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bumps log from 0.4.22 to 0.4.25.
Release notes
Sourced from log's releases.
Changelog
Sourced from log's changelog.
Commits
22be810Merge pull request #663 from rust-lang/cargo/0.4.250279730prepare for 0.4.25 release4099bcbMerge pull request #662 from rust-lang/fix/cargo-features36e7e3frevert loosening of kv cargo features2282191Merge pull request #660 from rust-lang/cargo/0.4.242994f0aprepare for 0.4.24 release5fcb50eMerge pull request #659 from rust-lang/fix/feature-builds29fe9e6fix up feature activationb1824f2use cargo hack in CI to test all feature combinationse6b643dMerge pull request #656 from rust-lang/cargo/0.4.23Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot mergewill merge this PR after your CI passes on it@dependabot squash and mergewill squash and merge this PR after your CI passes on it@dependabot cancel mergewill cancel a previously requested merge and block automerging@dependabot reopenwill reopen this PR if it is closed@dependabot closewill close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)