jarvisplot 2.1.7__tar.gz → 2.1.8__tar.gz

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/JarvisPLOT.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: JarvisPLOT
-Version: 2.1.7
+Version: 2.1.8
 Summary: Jarvis-PLOT: YAML-driven plotting engine
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -66,6 +66,7 @@ Render with a bare path — there is no `jplot run`:
 ```bash
 jplot path/to/config.yaml
 jplot path/to/config.yaml --rebuild-cache
+jplot path/to/config.yaml --print
 ```
 
 Everything else is discovery and judgement, in three groups.
@@ -218,9 +219,45 @@ from jarvisplot import render_flowchart, render_flowchart_file
 - You can set `project.workdir` in YAML.
 - If `output.dir` is omitted, Jarvis-PLOT defaults to `<workdir>/plots/`.
 - Data cache is stored in `<workdir>/.cache/`.
+- `--print` accepts the current settings and renders final figures without the logo. It announces that, after all enabled figures are saved successfully, cache payloads used or created by this run will be deleted (pipeline data and metadata, named/shared data, summaries, and HDF5 materialization slots). Rendering failures or a run with no enabled figures keep the cache. Unrelated cache entries and the lightweight fingerprint manifest are retained; later renders rebuild deleted entries as needed.
 - Profiling pipelines are prebuilt once and reused from cache when source fingerprint and profile settings are unchanged.
 - Profiling uses a fast two-stage grid reduction (`pregrid` + render `bin`) for large datasets.
 
+Plot caches are automatically registered in `~/.jarvis/plot.json` (or
+`$JARVIS_HOME/plot.json`). The registry groups caches by workdir and records
+associated YAML paths and last use; payloads stay in their original directories.
+This includes caches created by `data describe` and default `dryrun --with-data`
+exports. Sizes are calculated when listing, rather than stored in the registry.
+
+```bash
+jplot cache list
+jplot cache scan ~/Jarvis-Workshop               # register existing plot caches
+jplot cache clean --workdir /path/to/project    # this directory and all nested projects; repeatable
+jplot cache clean --all --older-than 30         # unused for at least 30 days
+jplot cache clean --all --dry-run               # preview; no deletion/index changes
+jplot cache clean --all                         # clean all registered inactive caches
+jplot cache -h
+jplot man cache
+```
+
+`--workdir PATH` selects registered caches in PATH itself and all of its
+subdirectories, even when PATH has no cache of its own. Repeat it to select
+multiple directory trees; overlapping selections are deduplicated. Use
+`cache scan PATH` first for old caches that have not been registered.
+
+All cache subcommands accept `--json` and use the existing CLI envelope and
+exit codes (0 success/partial, 1 failure, 2 usage). Non-TTY stdout defaults to
+JSON; human results go to stderr. Active caches are skipped (`ok: null` when
+the remaining operation succeeds), and records for missing paths are pruned
+during cleanup. Only marked Jarvis-PLOT components are removed; source data,
+plots, and unrelated `.cache` content are retained. Failed deletions retain
+their registry entries. Scanning adopts legacy caches only when the four
+pipeline directories and a compatible `manifest.json` identify them; old
+unmarked twin-only directories need a new `dryrun --with-data` run to register.
+An explicitly supplied `--out-dir` is an export directory and is not registered.
+Concurrent registry writes use a lock and atomic replacement. Registration
+failures warn without preventing rendering.
+
 ### Example: SUSYRun2 Ternary Plots
 
 ```bash

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/JarvisPLOT.egg-info/SOURCES.txt

@@ -9,6 +9,7 @@ JarvisPLOT.egg-info/top_level.txt
 jarvisplot/__init__.py
 jarvisplot/agent_digest.py
 jarvisplot/agent_io.py
+jarvisplot/cache_registry.py
 jarvisplot/cache_store.py
 jarvisplot/capabilities.py
 jarvisplot/cli.py
@@ -127,6 +128,7 @@ jarvisplot/cards/legends/default/legend-card.json
 jarvisplot/cards/legends/paper/legend-card.json
 jarvisplot/cards/legends/presentation/legend-card.json
 jarvisplot/manual_cards/agent-output.yaml
+jarvisplot/manual_cards/cache.yaml
 jarvisplot/manual_cards/cli-map.yaml
 jarvisplot/manual_cards/data-columns.yaml
 jarvisplot/manual_cards/debug-overlay.yaml
@@ -193,6 +195,7 @@ jarvisplot/utils/expression.py
 jarvisplot/utils/interpolator.py
 jarvisplot/utils/pathing.py
 jarvisplot/verbs/__init__.py
+jarvisplot/verbs/cache.py
 jarvisplot/verbs/cap.py
 jarvisplot/verbs/config_cmd.py
 jarvisplot/verbs/data.py
@@ -204,15 +207,19 @@ jarvisplot/verbs/suggest.py
 jarvisplot/verbs/template.py
 jarvisplot/verbs/validate.py
 tests/test_adaptive_cdf.py
+tests/test_add_column.py
 tests/test_agent_digest.py
 tests/test_agent_envelope.py
 tests/test_axis_title.py
+tests/test_cache_cleanup.py
+tests/test_cache_registry.py
 tests/test_capabilities.py
 tests/test_card_debug_shape.py
 tests/test_cli_help.py
 tests/test_cli_verbs.py
 tests/test_cmaps.py
 tests/test_column_existence.py
+tests/test_column_pruning.py
 tests/test_config_addressing.py
 tests/test_config_edit.py
 tests/test_config_expand.py
@@ -232,7 +239,9 @@ tests/test_diagnostic_guidance.py
 tests/test_distribution_1d_transform.py
 tests/test_docs_status_labels.py
 tests/test_doctor_partial.py
+tests/test_examples_validate.py
 tests/test_explain.py
+tests/test_expression_semantics.py
 tests/test_figure_contracts.py
 tests/test_figure_type_expander.py
 tests/test_fix_application.py
@@ -253,6 +262,7 @@ tests/test_pathing_contracts.py
 tests/test_posterior_hpd.py
 tests/test_profile_runtime.py
 tests/test_render_health.py
+tests/test_render_lifecycle.py
 tests/test_row_ledger.py
 tests/test_schema_catalog.py
 tests/test_suggest.py

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: JarvisPLOT
-Version: 2.1.7
+Version: 2.1.8
 Summary: Jarvis-PLOT: YAML-driven plotting engine
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -66,6 +66,7 @@ Render with a bare path — there is no `jplot run`:
 ```bash
 jplot path/to/config.yaml
 jplot path/to/config.yaml --rebuild-cache
+jplot path/to/config.yaml --print
 ```
 
 Everything else is discovery and judgement, in three groups.
@@ -218,9 +219,45 @@ from jarvisplot import render_flowchart, render_flowchart_file
 - You can set `project.workdir` in YAML.
 - If `output.dir` is omitted, Jarvis-PLOT defaults to `<workdir>/plots/`.
 - Data cache is stored in `<workdir>/.cache/`.
+- `--print` accepts the current settings and renders final figures without the logo. It announces that, after all enabled figures are saved successfully, cache payloads used or created by this run will be deleted (pipeline data and metadata, named/shared data, summaries, and HDF5 materialization slots). Rendering failures or a run with no enabled figures keep the cache. Unrelated cache entries and the lightweight fingerprint manifest are retained; later renders rebuild deleted entries as needed.
 - Profiling pipelines are prebuilt once and reused from cache when source fingerprint and profile settings are unchanged.
 - Profiling uses a fast two-stage grid reduction (`pregrid` + render `bin`) for large datasets.
 
+Plot caches are automatically registered in `~/.jarvis/plot.json` (or
+`$JARVIS_HOME/plot.json`). The registry groups caches by workdir and records
+associated YAML paths and last use; payloads stay in their original directories.
+This includes caches created by `data describe` and default `dryrun --with-data`
+exports. Sizes are calculated when listing, rather than stored in the registry.
+
+```bash
+jplot cache list
+jplot cache scan ~/Jarvis-Workshop               # register existing plot caches
+jplot cache clean --workdir /path/to/project    # this directory and all nested projects; repeatable
+jplot cache clean --all --older-than 30         # unused for at least 30 days
+jplot cache clean --all --dry-run               # preview; no deletion/index changes
+jplot cache clean --all                         # clean all registered inactive caches
+jplot cache -h
+jplot man cache
+```
+
+`--workdir PATH` selects registered caches in PATH itself and all of its
+subdirectories, even when PATH has no cache of its own. Repeat it to select
+multiple directory trees; overlapping selections are deduplicated. Use
+`cache scan PATH` first for old caches that have not been registered.
+
+All cache subcommands accept `--json` and use the existing CLI envelope and
+exit codes (0 success/partial, 1 failure, 2 usage). Non-TTY stdout defaults to
+JSON; human results go to stderr. Active caches are skipped (`ok: null` when
+the remaining operation succeeds), and records for missing paths are pruned
+during cleanup. Only marked Jarvis-PLOT components are removed; source data,
+plots, and unrelated `.cache` content are retained. Failed deletions retain
+their registry entries. Scanning adopts legacy caches only when the four
+pipeline directories and a compatible `manifest.json` identify them; old
+unmarked twin-only directories need a new `dryrun --with-data` run to register.
+An explicitly supplied `--out-dir` is an export directory and is not registered.
+Concurrent registry writes use a lock and atomic replacement. Registration
+failures warn without preventing rendering.
+
 ### Example: SUSYRun2 Ternary Plots
 
 ```bash

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/README.md

@@ -44,6 +44,7 @@ Render with a bare path — there is no `jplot run`:
 ```bash
 jplot path/to/config.yaml
 jplot path/to/config.yaml --rebuild-cache
+jplot path/to/config.yaml --print
 ```
 
 Everything else is discovery and judgement, in three groups.
@@ -196,9 +197,45 @@ from jarvisplot import render_flowchart, render_flowchart_file
 - You can set `project.workdir` in YAML.
 - If `output.dir` is omitted, Jarvis-PLOT defaults to `<workdir>/plots/`.
 - Data cache is stored in `<workdir>/.cache/`.
+- `--print` accepts the current settings and renders final figures without the logo. It announces that, after all enabled figures are saved successfully, cache payloads used or created by this run will be deleted (pipeline data and metadata, named/shared data, summaries, and HDF5 materialization slots). Rendering failures or a run with no enabled figures keep the cache. Unrelated cache entries and the lightweight fingerprint manifest are retained; later renders rebuild deleted entries as needed.
 - Profiling pipelines are prebuilt once and reused from cache when source fingerprint and profile settings are unchanged.
 - Profiling uses a fast two-stage grid reduction (`pregrid` + render `bin`) for large datasets.
 
+Plot caches are automatically registered in `~/.jarvis/plot.json` (or
+`$JARVIS_HOME/plot.json`). The registry groups caches by workdir and records
+associated YAML paths and last use; payloads stay in their original directories.
+This includes caches created by `data describe` and default `dryrun --with-data`
+exports. Sizes are calculated when listing, rather than stored in the registry.
+
+```bash
+jplot cache list
+jplot cache scan ~/Jarvis-Workshop               # register existing plot caches
+jplot cache clean --workdir /path/to/project    # this directory and all nested projects; repeatable
+jplot cache clean --all --older-than 30         # unused for at least 30 days
+jplot cache clean --all --dry-run               # preview; no deletion/index changes
+jplot cache clean --all                         # clean all registered inactive caches
+jplot cache -h
+jplot man cache
+```
+
+`--workdir PATH` selects registered caches in PATH itself and all of its
+subdirectories, even when PATH has no cache of its own. Repeat it to select
+multiple directory trees; overlapping selections are deduplicated. Use
+`cache scan PATH` first for old caches that have not been registered.
+
+All cache subcommands accept `--json` and use the existing CLI envelope and
+exit codes (0 success/partial, 1 failure, 2 usage). Non-TTY stdout defaults to
+JSON; human results go to stderr. Active caches are skipped (`ok: null` when
+the remaining operation succeeds), and records for missing paths are pruned
+during cleanup. Only marked Jarvis-PLOT components are removed; source data,
+plots, and unrelated `.cache` content are retained. Failed deletions retain
+their registry entries. Scanning adopts legacy caches only when the four
+pipeline directories and a compatible `manifest.json` identify them; old
+unmarked twin-only directories need a new `dryrun --with-data` run to register.
+An explicitly supplied `--out-dir` is an export directory and is not registered.
+Concurrent registry writes use a lock and atomic replacement. Registration
+failures warn without preventing rendering.
+
 ### Example: SUSYRun2 Ternary Plots
 
 ```bash

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/config_runtime.py

@@ -3,6 +3,8 @@ from __future__ import annotations
 from pathlib import Path
 from typing import Mapping
 
+from .style_runtime import normalize_style_tokens
+
 
 def apply_figure_config(fig, info: Mapping) -> bool:
     """Apply a YAML figure block to a Figure instance."""
@@ -55,7 +57,7 @@ def apply_figure_config(fig, info: Mapping) -> bool:
             )
 
         if "style" in info:
-            style_tokens = info["style"]
+            style_tokens = normalize_style_tokens(info["style"])
         else:
             style_tokens = ["a4paper_2x1"]
         fig.style = style_tokens

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/figure.py

@@ -11,7 +11,7 @@ from types import MethodType
 from .adapters_rect import StdAxesAdapter
 from .adapters_ternary import TernaryAxesAdapter
 from .method_registry import resolve_callable
-from .style_runtime import resolve_style_bundle_payload
+from .style_runtime import normalize_style_tokens, resolve_style_bundle_payload
 from .config_runtime import apply_figure_config
 from .design_runtime import draw_design_reference
 from .legend_runtime import apply_legend
@@ -399,10 +399,13 @@ class Figure:
     
     
     @style.setter
-    def style(self, value) -> None: 
-        if len(value) not in (1, 2):
-            self.logger.error("Undefined style -> {}".format(value))
-            raise TypeError
+    def style(self, value) -> None:
+        value = normalize_style_tokens(value)
+        if not isinstance(value, list) or len(value) not in (1, 2):
+            raise TypeError(
+                "style must name a card family or [family, variant], e.g. "
+                "a4paper_2x1 or [a4paper_2x1, rect]; got {!r}".format(value)
+            )
         family, selected, bundle = resolve_style_bundle_payload(self.jpstyles, value)
         self._frame = deepcopy(bundle["Frame"])
         self._legend_disabled_axes = set()
@@ -508,7 +511,7 @@ class Figure:
             return {kk: self._ensure_pandas_data(vv, reason=reason) for kk, vv in data.items()}
         return data
 
-    def _concat_loaded_data(self, items):
+    def _concat_loaded_data(self, items, layer_name: str = ""):
         if len(items) == 0:
             return None
         if len(items) == 1:
@@ -519,7 +522,28 @@ class Figure:
         pandas_items = [self._ensure_pandas_data(item, reason="concat-layer") for item in items]
         try:
             return pd.concat(pandas_items, ignore_index=False)
-        except Exception:
+        except Exception as exc:
+            # Keep drawing what can be drawn, but never drop data silently.
+            if self.logger:
+                shapes = ", ".join(
+                    str(getattr(item, "shape", type(item).__name__)) for item in pandas_items
+                )
+                self.logger.warning(
+                    "Layer '{}': the {} data blocks could not be concatenated, so only "
+                    "the first block is drawn and the other {} are dropped.\n"
+                    "\t reason \t-> {}: {}\n"
+                    "\t blocks \t-> {}\n"
+                    "\t hint \t\t-> blocks combined with `combine: concat` must be tables "
+                    "with compatible columns (no duplicated column names); use "
+                    "`combine: separate` to draw them independently.".format(
+                        layer_name or "<unnamed>",
+                        len(pandas_items),
+                        len(pandas_items) - 1,
+                        type(exc).__name__,
+                        exc,
+                        shapes,
+                    )
+                )
             return pandas_items[0]
 
     def _store_share_data_if_needed(self, layer, data, cache_ref: str | None = None):

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/layer_runtime.py

@@ -642,7 +642,7 @@ def _load_layer_data(fig, layer):
                     fig.logger.error("DataSet -> {} not specified".format(src))
             if len(dts) == 0:
                 return None, None
-            combined = fig._concat_loaded_data(dts)
+            combined = fig._concat_loaded_data(dts, layer_name=str(layer.get("name", "") or ""))
             cache_ref = None
             if len(dts) == 1 and combined is dts[0] and len(cache_keys) == 1:
                 cache_ref = cache_keys[0]

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/preprocessor.py

@@ -883,6 +883,7 @@ class DataPreprocessor:
             "mode": str(mode),
             "projection": self._projection_list(projection),
         }
+        self._add_source_column_scope(payload, source, projection)
         if self.cache is not None:
             return self.cache.cache_key(payload)
         return self._stable_hash(payload)
@@ -896,7 +897,7 @@ class DataPreprocessor:
         projection: Optional[Sequence[str]] = None,
     ) -> Dict[str, Any]:
         eff_transform = self._effective_transform(source, transform)
-        return {
+        payload = {
             "schema": "jp-demand-v8",
             "source": self._source_token(source, combine=combine),
             "extra_sources": self._transform_extra_source_tokens(eff_transform),
@@ -905,6 +906,29 @@ class DataPreprocessor:
             "mode": str(mode),
             "projection": self._projection_list(projection),
         }
+        self._add_source_column_scope(payload, source, projection)
+        return payload
+
+    def _add_source_column_scope(self, payload: Dict[str, Any], source: Any, projection) -> None:
+        """Key an unprojected pipeline on the columns its datasets were loaded with.
+
+        A projected pipeline keeps the same columns whatever the column plan
+        loaded. An unprojected one (a published table, a step that takes the
+        whole table) carries every loaded column, so a result cached under one
+        plan would be missing columns a later config asks for. The scope is only
+        added when a dataset is actually pruned, so other keys stay as they were.
+        """
+        if self._projection_list(projection) is not None:
+            return
+        names = [source] if isinstance(source, str) else list(source) if isinstance(source, (list, tuple)) else []
+        scope = {}
+        for name in names:
+            dts = self.dataset_registry.get(str(name))
+            planned = getattr(dts, "required_columns", None) if dts is not None else None
+            if planned:
+                scope[str(name)] = sorted(planned)
+        if scope:
+            payload["source_columns"] = scope
 
     def _demand_fingerprint(
         self,
@@ -935,6 +959,9 @@ class DataPreprocessor:
                 profile_sig = self._runtime_profile_signature(tf)
                 if profile_sig is not None:
                     token["profile_signature"] = profile_sig
+                self._add_source_column_scope(
+                    token, src, self._runtime_projection(tf, self.layer_demand_columns(layer))
+                )
                 tokens.append(token)
         payload = {
             "name": layer.get("name"),

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/preprocessor_runtime.py

@@ -235,7 +235,7 @@ def filter_df(df, condition, logger):
                 return df.copy(deep=False)
             if low in {"false", "f", "no", "n"}:
                 return df.iloc[0:0].copy()
-            s = s.replace("&&", " & ").replace("||", " | ")
+            # `&&` / `||` / and / or / not are handled by the shared parser.
             condition = s
         else:
             raise TypeError(f"Unsupported condition type: {type(condition)}")
@@ -255,13 +255,35 @@ def filter_df(df, condition, logger):
         return pd.DataFrame(index=df.index).iloc[0:0].copy()
 
 
+def add_column_spec_problem(adds) -> Optional[str]:
+    """Why an ``add_column`` step cannot run, or ``None`` when it can."""
+    if not isinstance(adds, Mapping):
+        return f"add_column skipped: expected {{name, expr}}, got {adds!r}"
+    missing = []
+    if not str(adds.get("name") or "").strip():
+        missing.append("name")
+    expr = adds.get("expr")
+    if expr is None or not str(expr).strip():
+        missing.append("expr")
+    if missing:
+        return "add_column skipped: missing {} -> {}".format(" and ".join(missing), dict(adds))
+    return None
+
+
 def add_column(df, adds, logger):
     try:
-        name = adds.get("name", False)
-        expr = adds.get("expr", False)
-        if not (name and expr):
-            logger.error("Error in loading add_column -> {}".format(adds))
+        problem = add_column_spec_problem(adds)
+        if problem:
+            # Nothing sensible to add: carry on without the column rather than
+            # inventing one (an absent expr used to evaluate to False).
+            logger.error(problem)
+            return df
+        name = str(adds.get("name")).strip()
+        expr = adds.get("expr")
         value = eval_dataframe_expression(df, expr, logger=logger, allow_column=True)
+        # The frame may be a table other blocks still read (a preprofile pass
+        # works on the source itself); add to a copy, never into the caller's.
+        df = df.copy(deep=False)
         df[name] = value
         return df
     except Exception as e:

{jarvisplot-2.1.7 → jarvisplot-2.1.8}/jarvisplot/Figure/style_runtime.py

@@ -8,8 +8,22 @@ from typing import Any, Mapping, Tuple
 PREFERRED_STYLE_VARIANTS = ("rect", "Ternary", "rect_5x1", "dynesty_runplot", "rectcmap", "TernaryCmap")
 
 
+def normalize_style_tokens(value):
+    """``style: a4paper_2x1`` and ``style: [a4paper_2x1]`` name the same card.
+
+    The schema accepts both spellings; everything downstream indexes tokens,
+    and a bare string would be read one character at a time.
+    """
+    if isinstance(value, str):
+        return [value]
+    if isinstance(value, tuple):
+        return list(value)
+    return value
+
+
 def resolve_style_bundle_payload(jpstyles: Mapping[str, Any], value) -> Tuple[str, str, dict]:
     """Resolve style tokens to a full bundle payload."""
+    value = normalize_style_tokens(value)
     if len(value) == 2:
         family_name = value[0]
         variant_name = value[1]