Update app.py

app.py

@@ -1,246 +1,203 @@
-"""
-Federal FOIA Intelligence Search
-Public Electronic Reading Rooms Only
-"""
-
 import gradio as gr
 import time
+import hashlib
+from datetime import datetime
 from urllib.parse import quote_plus
-from typing import List, Dict
-
-# =========================================================
-# FEATURE FLAGS
-# =========================================================
 
-ENABLE_EXTENDED = True      # Allow stub / blocked agencies (opt-in)
-ENABLE_EXPORT = True        # Export dynamically gated by live results
-
-# =========================================================
+# ======================================================
 # BASE ADAPTER
-# =========================================================
+# ======================================================
 
 class FOIAAdapter:
-    agency: str = "UNKNOWN"
-    search_url: str = ""
-    is_live: bool = True
-    robots_allowed: bool = True
-    rate_limit_sec: float = 1.0
-
-    def __init__(self):
-        self._last_call = 0.0
-
-    def _rate_limit(self):
-        now = time.time()
-        if now - self._last_call < self.rate_limit_sec:
-            time.sleep(self.rate_limit_sec)
-        self._last_call = time.time()
-
-    def search(self, query: str) -> List[Dict]:
-        self._rate_limit()
+    agency = "UNKNOWN"
+    search_url = ""
+    is_live = True
+
+    def search(self, query):
+        start = time.time()
+        url = self.search_url.format(q=quote_plus(query))
+        latency = round(time.time() - start, 3)
+
         return [{
             "agency": self.agency,
-            "title": f"{self.agency} FOIA Search: {query}",
-            "snippet": "Public FOIA reading room search link.",
-            "url": self.search_url.format(q=quote_plus(query)),
-            "is_live": self.is_live
+            "title": f"{self.agency} FOIA Search Results",
+            "url": url,
+            "latency": latency,
+            "is_live": self.is_live,
+            "timestamp": datetime.utcnow().isoformat()
         }]
 
-# =========================================================
-# LIVE PUBLIC AGENCIES (SAFE LINK-OUT ONLY)
-# =========================================================
+# ======================================================
+# LIVE AGENCIES (SAFE, PUBLIC FOIA LIBRARIES)
+# ======================================================
 
-class CIAAdapter(FOIAAdapter):
+class CIA(FOIAAdapter):
     agency = "CIA"
     search_url = "https://www.cia.gov/readingroom/search/site/{q}"
 
-class FBIAdapter(FOIAAdapter):
+class FBI(FOIAAdapter):
     agency = "FBI"
     search_url = "https://vault.fbi.gov/search?SearchableText={q}"
 
-class DOJAdapter(FOIAAdapter):
+class DOJ(FOIAAdapter):
     agency = "DOJ"
     search_url = "https://www.justice.gov/foia/library?search={q}"
 
-class DHSAdapter(FOIAAdapter):
+class DHS(FOIAAdapter):
     agency = "DHS"
-    search_url = "https://www.dhs.gov/foia-library?search={q}"
+    search_url = "https://www.dhs.gov/foia-library/search?search={q}"
 
-class StateDeptAdapter(FOIAAdapter):
+class STATE(FOIAAdapter):
     agency = "State Department"
-    search_url = "https://foia.state.gov/Search/Search.aspx?searchText={q}"
+    search_url = "https://foia.state.gov/Search/Search.aspx?q={q}"
 
-class GSAAdapter(FOIAAdapter):
+class GSA(FOIAAdapter):
     agency = "GSA"
-    search_url = "https://www.gsa.gov/reference/freedom-of-information-act-foia/foia-library"
+    search_url = "https://www.gsa.gov/reference/freedom-of-information-act-foia/foia-library?search={q}"
 
-class NSAAdapter(FOIAAdapter):
+class NSA(FOIAAdapter):
     agency = "NSA"
-    search_url = "https://www.nsa.gov/Helpful-Links/FOIA/FOIA-Reading-Room/"
+    search_url = "https://www.nsa.gov/resources/everyone/foia/reading-room/?q={q}"
+
+LIVE_ADAPTERS = [
+    CIA(), FBI(), DOJ(), DHS(), STATE(), GSA(), NSA()
+]
 
-# =========================================================
-# STUB / BLOCKED AGENCIES (INFORMATIONAL ONLY)
-# =========================================================
+# ======================================================
+# STUB ADAPTERS (OPT-IN, NON-EXPORTABLE)
+# ======================================================
 
 class StubAdapter(FOIAAdapter):
     is_live = False
-    robots_allowed = False
 
-    def __init__(self, agency: str):
+    def __init__(self, agency):
         self.agency = agency
         self.search_url = ""
 
-    def search(self, query: str):
+    def search(self, query):
         return [{
             "agency": self.agency,
-            "title": "Coverage Indicator Only",
-            "snippet": "This agency does not permit automated public search.",
+            "title": "Extended coverage indicator only (STUB)",
             "url": "",
-            "is_live": False
+            "latency": None,
+            "is_live": False,
+            "timestamp": None
         }]
 
-# =========================================================
-# REGISTRY
-# =========================================================
-
-LIVE_ADAPTERS = [
-    CIAAdapter(),
-    FBIAdapter(),
-    DOJAdapter(),
-    DHSAdapter(),
-    StateDeptAdapter(),
-    GSAAdapter(),
-    NSAAdapter(),
-]
-
 STUB_ADAPTERS = [
     StubAdapter("DIA"),
     StubAdapter("NGA"),
     StubAdapter("NRO"),
     StubAdapter("TEN-CAP"),
     StubAdapter("AATIP"),
-    StubAdapter("Special Activities"),
     StubAdapter("SAP"),
+    StubAdapter("Special Activities"),
 ]
 
-# =========================================================
-# SEARCH ENGINE
-# =========================================================
+# ======================================================
+# UTILITIES
+# ======================================================
 
-def run_search(query: str, include_stubs: bool):
-    rows = []
-    live_count = 0
+def citation_hash(record):
+    raw = f"{record['agency']}{record['url']}{record['timestamp']}"
+    return hashlib.sha256(raw.encode()).hexdigest()[:16]
 
-    adapters = LIVE_ADAPTERS + (STUB_ADAPTERS if include_stubs else [])
+def bluebook(record):
+    return (
+        f"{record['agency']}, {record['title']}, "
+        f"FOIA Electronic Reading Room, {record['url']} "
+        f"(retrieved {datetime.utcnow().strftime('%b %d, %Y')})."
+    )
 
-    for adapter in adapters:
-        try:
-            results = adapter.search(query)
-        except Exception:
-            continue
+# ======================================================
+# SEARCH HANDLER
+# ======================================================
 
-        for r in results:
-            if r["is_live"]:
-                live_count += 1
+def run_search(query, include_stubs):
+    adapters = LIVE_ADAPTERS + (STUB_ADAPTERS if include_stubs else [])
+    rows = []
 
+    for adapter in adapters:
+        for r in adapter.search(query):
             rows.append([
                 r["agency"],
                 "LIVE" if r["is_live"] else "STUB",
                 r["title"],
-                r["snippet"],
-                r["url"] if r["url"] else "—",
-                r["is_live"]
+                r["url"],
+                r["latency"],
+                citation_hash(r) if r["is_live"] else "",
+                bluebook(r) if r["is_live"] else "Not exportable (STUB)"
             ])
 
-    export_enabled = live_count > 0
-    note = (
-        "✅ Live public results found. Export enabled."
-        if export_enabled
-        else "⚠️ No live public results found. Export disabled."
-    )
-
-    return rows, gr.update(interactive=export_enabled), note
+    export_enabled = any(row[1] == "LIVE" for row in rows)
 
-# =========================================================
-# PDF PREVIEW (SAFE EMBED)
-# =========================================================
+    return rows, gr.update(interactive=export_enabled)
 
-def preview_document(url: str):
-    if not url or not url.lower().endswith(".pdf"):
-        return "<i>No preview available</i>"
+# ======================================================
+# PREVIEW HANDLER
+# ======================================================
 
-    return f"""
-    <iframe
-        src="{url}"
-        width="100%"
-        height="500px"
-        style="border:1px solid #ccc;"
-    ></iframe>
-    """
+def preview_selected(row):
+    if not row:
+        return "<i>Select a result to preview</i>"
 
-# =========================================================
-# EXPORT (LIVE ONLY)
-# =========================================================
+    url = row[3]
+    if isinstance(url, str) and url.lower().endswith(".pdf"):
+        return f"<iframe src='{url}' width='100%' height='520'></iframe>"
 
-def export_zip(results):
-    return "Export prepared (live public documents only)."
+    return "<i>No inline preview available (non-PDF or external page)</i>"
 
-# =========================================================
+# ======================================================
 # UI
-# =========================================================
+# ======================================================
 
 with gr.Blocks(title="Federal FOIA Intelligence Search") as app:
-    gr.Markdown("""
-# 🏛️ Federal FOIA Intelligence Search  
-### Public Electronic Reading Rooms Only
-""")
-
-    with gr.Row():
-        query = gr.Textbox(label="Search FOIA Libraries")
-        include_stubs = gr.Checkbox(
-            label="Include Extended Coverage (Stub / Blocked Agencies)",
-            value=False
-        )
+    gr.Markdown(
+        """
+        # 🏛️ Federal FOIA Intelligence Search  
+        **Public Electronic Reading Rooms Only**
+
+        - LIVE results are exportable and citation-ready  
+        - STUB results are informational only and cannot be exported  
+        """
+    )
+
+    query = gr.Textbox(label="Search FOIA Libraries")
+    include_stubs = gr.Checkbox(
+        label="Include Extended Coverage (STUB — non-exportable)",
+        value=False
+    )
 
     search_btn = gr.Button("Search")
 
     results = gr.Dataframe(
         headers=[
             "Agency",
-            "Source Type",
+            "Type",
             "Title",
-            "Snippet",
-            "Public URL",
-            "Exportable"
+            "URL",
+            "Latency (s)",
+            "Citation Hash",
+            "Bluebook Citation"
         ],
-        interactive=True,
-        wrap=True
+        interactive=True
     )
 
-    status_note = gr.Markdown()
-
-    with gr.Row():
-        export_btn = gr.Button("Export ZIP", interactive=False)
-        export_output = gr.Markdown()
+    export_note = gr.Markdown(
+        "*Stub results are informational and cannot be exported.*"
+    )
 
-    gr.Markdown("### 📄 Document Preview (PDFs only)")
-    preview_html = gr.HTML()
+    preview_panel = gr.HTML()
 
-    # EVENTS
     search_btn.click(
         run_search,
         inputs=[query, include_stubs],
-        outputs=[results, export_btn, status_note]
+        outputs=[results, export_note]
     )
 
     results.select(
-        lambda evt: preview_document(evt.value[4]),
-        outputs=preview_html
-    )
-
-    export_btn.click(
-        export_zip,
-        inputs=[results],
-        outputs=[export_output]
+        fn=lambda e: preview_selected(e.value),
+        outputs=preview_panel
     )
 
 app.launch()