feat: persist workbook home and add dashboard manuscript intake bridge

.gitattributes

@@ -37,6 +37,3 @@ smoke_signal/tessdata/gill_sans_infant.traineddata.otf filter=lfs diff=lfs merge
 *.traineddata filter=lfs diff=lfs merge=lfs -text
 *.otf filter=lfs diff=lfs merge=lfs -text
 assets/Brand/Side[[:space:]]Navigation[[:space:]]Bar.png filter=lfs diff=lfs merge=lfs -text
-assets/Brand/Ui[[:space:]]Totem.png filter=lfs diff=lfs merge=lfs -text
-assets/Brand/Hero.png filter=lfs diff=lfs merge=lfs -text
-assets/Brand/side_nav_logo.png filter=lfs diff=lfs merge=lfs -text

app.py

@@ -4,6 +4,7 @@ import hashlib
 import html
 import json
 import re
+from shutil import copy2
 from html import escape
 from pathlib import Path
 
@@ -31,6 +32,8 @@ from smoke_signal_tab import smoke_signal_tab, SS_CSS
 
 ORIGINAL_WORKBOOK_PATH = "data/order69_macmillan_totem_rebuilt.xlsx"
 CODEX_CATALOGUE_PATH = Path("data/codex_catalogue.xlsx")
+WORKBOOK_HOME_DIR = Path("data/workbook_home")
+ACTIVE_WORKBOOK_PATH = WORKBOOK_HOME_DIR / "active_workbook.xlsx"
 
 
 def _patch_gradio_schema_bool_compat() -> None:
@@ -1794,6 +1797,32 @@ def _validate_workbook_path(path: Path) -> Path:
     return path
 
 
+def _ensure_workbook_home() -> None:
+    WORKBOOK_HOME_DIR.mkdir(parents=True, exist_ok=True)
+
+
+def _persist_uploaded_workbook(uploaded_file) -> Path:
+    """
+    Copy workbook upload to a stable local 'home' path so users do not need
+    to re-upload each session.
+    """
+    src = _validate_workbook_path(_clean_path(uploaded_file))
+    _ensure_workbook_home()
+    copy2(src, ACTIVE_WORKBOOK_PATH)
+    return ACTIVE_WORKBOOK_PATH
+
+
+def _preferred_workbook_path() -> Path:
+    """
+    Resolve workbook 'home' priority:
+    1) persisted uploaded workbook
+    2) bundled default workbook
+    """
+    if ACTIVE_WORKBOOK_PATH.exists():
+        return ACTIVE_WORKBOOK_PATH
+    return DEFAULT_WORKBOOK
+
+
 def _score_summary(log_df: pd.DataFrame | None) -> str:
     if log_df is None or log_df.empty:
         return "No scored rows yet."
@@ -2685,13 +2714,14 @@ def load_workbook(uploaded_file=None, notice: str = ""):
 
 
 def load_default():
-    return load_workbook(None, "Bundled workbook reloaded.")
+    return load_workbook(_preferred_workbook_path(), "Workbook reloaded from persistent home.")
 
 
 def load_uploaded(uploaded_file):
     if uploaded_file is None:
         raise gr.Error("Choose an .xlsx or .xlsm workbook first.")
-    return load_workbook(uploaded_file, "Workbook uploaded and analysed.")
+    persisted = _persist_uploaded_workbook(uploaded_file)
+    return load_workbook(persisted, f"Workbook uploaded and saved to {persisted}.")
 
 
 def load_local_path(path_text: str):
@@ -2702,27 +2732,76 @@ def load_local_path(path_text: str):
 
 
 def run_analysis(active_path: str):
-    path = _validate_workbook_path(Path(active_path) if active_path else DEFAULT_WORKBOOK)
+    path = _validate_workbook_path(Path(active_path) if active_path else _preferred_workbook_path())
     log_df = score_log(path)
     state = _build_dashboard_state_from_workbook(path, "complete", "TOTEM analysis refreshed.")
     return render_dashboard(state), log_df, _score_summary(log_df)
 
 
+def analyze_manuscript_and_refresh(
+    file_obj,
+    author_name: str,
+    author_id: str,
+    works_sampled: str,
+    active_path: str,
+):
+    """
+    Dashboard bridge:
+    - run manuscript extraction
+    - refresh workbook dashboard view so scoring context stays synced
+    """
+    report, json_out, status = run_codex_extraction(file_obj, author_name, author_id, works_sampled)
+    try:
+        dashboard_html, log_df, summary = run_analysis(active_path)
+    except Exception:
+        dashboard_html = initial_dashboard_html()
+        log_df = pd.DataFrame(columns=LOG_COLUMNS)
+        summary = "Workbook refresh skipped because active workbook path is invalid."
+    return report, json_out, status, dashboard_html, log_df, summary
+
+
+def analyze_manuscript_for_dashboard(file_obj, active_path: str):
+    """
+    Minimal dashboard-facing manuscript pipeline:
+    - accepts a manuscript upload directly on Dashboard
+    - runs Codex extraction
+    - refreshes workbook dashboard state
+    """
+    file_path = _extract_uploaded_path(file_obj)
+    if not file_path:
+        raise gr.Error("Upload a manuscript (.txt or .pdf) first.")
+
+    stem = Path(file_path).stem.strip() or "Unknown Author"
+    author_name = stem.replace("_", " ")[:120]
+    works_sampled = stem.replace("_", " ")[:180]
+    author_id = "CA-XXX"
+    report, json_out, status, dashboard_html, log_df, summary = analyze_manuscript_and_refresh(
+        file_obj=file_obj,
+        author_name=author_name,
+        author_id=author_id,
+        works_sampled=works_sampled,
+        active_path=active_path,
+    )
+    # Keep dashboard status compact while preserving full report in dedicated output.
+    dashboard_status = f"{status} | Workbook refresh: {summary}"
+    return report, json_out, dashboard_status, dashboard_html, log_df, summary
+
+
 def recalc_log(log_df, active_path: str):
-    path = _validate_workbook_path(Path(active_path) if active_path else DEFAULT_WORKBOOK)
+    path = _validate_workbook_path(Path(active_path) if active_path else _preferred_workbook_path())
     recalculated = recalculate_log(log_df, path)
     state = _build_dashboard_state_from_workbook(path, "complete", "Gates recalculated.")
     return render_dashboard(state), recalculated, _score_summary(recalculated)
 
 
 def export_log(log_df, active_path: str):
-    path = _validate_workbook_path(Path(active_path) if active_path else DEFAULT_WORKBOOK)
+    path = _validate_workbook_path(Path(active_path) if active_path else _preferred_workbook_path())
     return export_updated_workbook(log_df, path)
 
 
 def single_score(active_path, sequence, stanza_id, draft_pass, clarity, rhythm, flow,
                  emotional_truth, visual_strength, commercial, notes):
-    path = _validate_workbook_path(Path(active_path) if active_path else DEFAULT_WORKBOOK)
+    path = _validate_workbook_path(Path(active_path) if active_path else _preferred_workbook_path())
     df = score_single_row(path, sequence, stanza_id, draft_pass, clarity, rhythm,
                           flow, emotional_truth, visual_strength, commercial, notes)
     return df, _score_summary(df)
@@ -2734,7 +2813,7 @@ def initial_dashboard_html() -> str:
     Falls back gracefully if workbook read fails.
     """
     state = get_initial_dashboard_state()
-    state["workbook_loaded"] = bool(DEFAULT_WORKBOOK.exists())
+    state["workbook_loaded"] = bool(_preferred_workbook_path().exists())
     state["analysis_status"] = "ready" if state["workbook_loaded"] else "idle"
     state["project_name"] = "Editorial Workspace"
     state["totem_signal"] = compute_totem_signal(
@@ -2748,7 +2827,7 @@ def initial_dashboard_html() -> str:
 # ── GRADIO INTERFACE ──────────────────────────────────────────────────────────
 
 with gr.Blocks(title="TOTEM Studio") as demo:
-    active_path = gr.State(str(DEFAULT_WORKBOOK))
+    active_path = gr.State(str(_preferred_workbook_path()))
     log_state = gr.State(pd.DataFrame(columns=LOG_COLUMNS))
 
     with gr.Tabs():
@@ -2768,9 +2847,40 @@ with gr.Blocks(title="TOTEM Studio") as demo:
                     )
                     run_button = gr.Button("Run TOTEM Analysis", variant="secondary", scale=1)
 
+            with gr.Accordion("Manuscript Intake (score pipeline)", open=False, elem_id="codex-panel"):
+                with gr.Column(elem_classes=["codex-body"]):
+                    gr.Markdown(
+                        "Upload manuscript here to run extraction and refresh workbook scoring context."
+                    )
+                    dashboard_manuscript_file = gr.File(
+                        label="Manuscript (.txt or .pdf)",
+                        file_types=[".txt", ".pdf"],
+                        type="filepath",
+                    )
+                    dashboard_manuscript_analyze = gr.Button(
+                        "Analyze Manuscript + Refresh Dashboard",
+                        variant="primary",
+                    )
+                    dashboard_manuscript_status = gr.Textbox(
+                        label="Pipeline status",
+                        interactive=False,
+                        lines=2,
+                        value="Ready.",
+                    )
+                    dashboard_manuscript_json = gr.Textbox(
+                        label="Extracted fingerprint (JSON)",
+                        interactive=False,
+                        lines=8,
+                    )
+                    dashboard_manuscript_report = gr.Textbox(
+                        label="Extraction report",
+                        interactive=False,
+                        lines=10,
+                    )
+
             with gr.Row(elem_id="path-panel"):
                 with gr.Column(elem_classes=["wrap"]):
-                    path_input = gr.Textbox(label="Local workbook path", value=ORIGINAL_WORKBOOK_PATH)
+                    path_input = gr.Textbox(label="Local workbook path", value=str(_preferred_workbook_path()))
                     path_button = gr.Button("Load Local Path", variant="primary")
 
             with gr.Accordion("Private scoring controls", open=False, elem_id="score-panel"):
@@ -2926,6 +3036,18 @@ with gr.Blocks(title="TOTEM Studio") as demo:
                      outputs=[dashboard, log_state, score_status])
     path_button.click(load_local_path, inputs=[path_input],
                       outputs=[active_path, dashboard, log_state, score_status])
+    dashboard_manuscript_analyze.click(
+        analyze_manuscript_for_dashboard,
+        inputs=[dashboard_manuscript_file, active_path],
+        outputs=[
+            dashboard_manuscript_report,
+            dashboard_manuscript_json,
+            dashboard_manuscript_status,
+            dashboard,
+            log_state,
+            score_status,
+        ],
+    )
     recalc_button.click(recalc_log, inputs=[log_state, active_path],
                         outputs=[dashboard, log_state, score_status])
     export_button.click(export_log, inputs=[log_state, active_path], outputs=[exported_file])