Upload folder using huggingface_hub

builders/build_docx.py

@@ -158,6 +158,18 @@ def _swot_section(doc, data):
         ("Opportunities", "opportunities", "DDEBF7", NAVY),
         ("Threats",       "threats",       "FFEB9C", RGBColor(0x7F,0x60,0x00)),
     ]
+    
+    parsed_any = any(len(swot.get(k, [])) > 0 for _, k, _, _ in configs)
+    
+    if not parsed_any and swot.get("raw"):
+        for line in swot["raw"].split("\n"):
+            if line.strip():
+                p = doc.add_paragraph()
+                r = p.add_run(line.strip())
+                r.font.size = Pt(11); r.font.name = "Calibri"
+        doc.add_paragraph()
+        return
+        
     t = doc.add_table(rows=2, cols=2)
     t.alignment = WD_TABLE_ALIGNMENT.LEFT
     idx = 0

builders/build_excel.py

@@ -394,6 +394,138 @@ def _swot_sheet(wb, data):
     _widths(ws, [90])
 
 
+def _dcf_sheet(wb, data):
+    ws = wb.create_sheet("DCF Valuation")
+    _title(ws, "Discounted Cash Flow (DCF) Model", "Interactive Valuation Model")
+    
+    co = data.get("company", {})
+    cf = data.get("cash_flow", {})
+    inc = data.get("income_statement", {})
+    bal = data.get("balance_sheet", {})
+    
+    # Get latest data
+    fcf_history = [v for v in cf.get("free_cash_flow", []) if v is not None]
+    latest_fcf = fcf_history[-1] if fcf_history else 0
+    
+    shares_history = [v for v in inc.get("diluted_shares", []) if v is not None]
+    latest_shares = shares_history[-1] if shares_history else 1
+    
+    cash_history = [v for v in bal.get("cash_and_equivalents", []) if v is not None]
+    cash = cash_history[-1] if cash_history else 0
+    
+    debt_history = [v for v in bal.get("total_liabilities", []) if v is not None]
+    debt = debt_history[-1] if debt_history else 0
+    
+    # Inputs
+    r = 4
+    _sec(ws, r, "Model Inputs (Editable - Yellow Cells)", 2); r += 1
+    
+    ws.cell(row=r, column=1, value="Starting FCF (Base)").font = LBL_FONT
+    ws.cell(row=r, column=2, value=latest_fcf).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    ws.cell(row=r, column=1, value="WACC (Discount Rate)").font = LBL_FONT
+    ws.cell(row=r, column=2, value=0.10).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_PCT; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    ws.cell(row=r, column=1, value="Terminal Growth Rate").font = LBL_FONT
+    ws.cell(row=r, column=2, value=0.025).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_PCT; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    ws.cell(row=r, column=1, value="Cash & Equivalents").font = LBL_FONT
+    ws.cell(row=r, column=2, value=cash).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    ws.cell(row=r, column=1, value="Total Debt").font = LBL_FONT
+    ws.cell(row=r, column=2, value=debt).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    ws.cell(row=r, column=1, value="Diluted Shares").font = LBL_FONT
+    ws.cell(row=r, column=2, value=latest_shares).font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="FFFFE0"); r += 1
+    
+    r += 1
+    _sec(ws, r, "Projected Cash Flows (5 Years)", 6); r += 1
+    _hdr(ws, r, ["Year 1", "Year 2", "Year 3", "Year 4", "Year 5"], start=2); r += 1
+    
+    # Growth rates
+    ws.cell(row=r, column=1, value="Projected Growth Rate").font = LBL_FONT
+    for c in range(2, 7):
+        ws.cell(row=r, column=c, value=0.05).font = DAT_FONT; ws.cell(row=r, column=c).number_format = FMT_PCT; ws.cell(row=r, column=c).fill = PatternFill("solid", fgColor="FFFFE0")
+    r += 1
+    
+    # FCF Projection Formulas
+    ws.cell(row=r, column=1, value="Projected FCF").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=B5*(1+B14)") # Year 1 = Base * (1+g1)
+    ws.cell(row=r, column=3, value="=B15*(1+C14)")
+    ws.cell(row=r, column=4, value="=C15*(1+D14)")
+    ws.cell(row=r, column=5, value="=D15*(1+E14)")
+    ws.cell(row=r, column=6, value="=E15*(1+F14)")
+    for c in range(2, 7):
+        ws.cell(row=r, column=c).font = DAT_FONT; ws.cell(row=r, column=c).number_format = FMT_MONEY
+    r += 1
+    
+    # Discount Factor Formulas
+    ws.cell(row=r, column=1, value="Discount Factor").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=1/((1+$B$6)^1)")
+    ws.cell(row=r, column=3, value="=1/((1+$B$6)^2)")
+    ws.cell(row=r, column=4, value="=1/((1+$B$6)^3)")
+    ws.cell(row=r, column=5, value="=1/((1+$B$6)^4)")
+    ws.cell(row=r, column=6, value="=1/((1+$B$6)^5)")
+    for c in range(2, 7):
+        ws.cell(row=r, column=c).font = DAT_FONT; ws.cell(row=r, column=c).number_format = FMT_RATIO
+    r += 1
+    
+    # PV of FCF Formulas
+    ws.cell(row=r, column=1, value="PV of FCF").font = LBL_FONT
+    for c, col_l in enumerate(["B", "C", "D", "E", "F"], start=2):
+        ws.cell(row=r, column=c, value=f"={col_l}15*{col_l}16").font = DAT_FONT; ws.cell(row=r, column=c).number_format = FMT_MONEY
+    r += 2
+    
+    # Valuation Output
+    _sec(ws, r, "Valuation Output", 2); r += 1
+    
+    ws.cell(row=r, column=1, value="Sum of PV of FCF").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=SUM(B17:F17)").font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; r += 1
+    
+    ws.cell(row=r, column=1, value="Terminal Value").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=(F15*(1+B7))/(B6-B7)").font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; r += 1
+    
+    ws.cell(row=r, column=1, value="PV of Terminal Value").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=B21*F16").font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; r += 1
+    
+    ws.cell(row=r, column=1, value="Implied Enterprise Value").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=B20+B22").font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; r += 1
+    
+    ws.cell(row=r, column=1, value="Implied Equity Value").font = LBL_FONT
+    ws.cell(row=r, column=2, value="=B23+B8-B9").font = DAT_FONT; ws.cell(row=r, column=2).number_format = FMT_MONEY; r += 1
+    
+    ws.cell(row=r, column=1, value="Implied Share Price").font = Font(name="Calibri", size=12, bold=True, color="1F4E78")
+    ws.cell(row=r, column=2, value="=B24/B10").font = Font(name="Calibri", size=12, bold=True); ws.cell(row=r, column=2).number_format = FMT_DEC; ws.cell(row=r, column=2).fill = PatternFill("solid", fgColor="DDEBF7")
+    
+    _widths(ws, [30, 20, 20, 20, 20, 20])
+
+
+def _narrative_sheet(wb, data):
+    ws = wb.create_sheet("Industry & AI Insights")
+    _title(ws, "AI Narrative Analysis", "Generated by AI Models")
+    r = 4
+    
+    exec_summary = data.get("executive_summary", "")
+    if exec_summary:
+        _sec(ws, r, "Executive Summary", 2); r += 1
+        c = ws.cell(row=r, column=1, value=exec_summary)
+        c.font = LBL_FONT; c.alignment = Alignment(wrap_text=True, vertical="top")
+        ws.merge_cells(start_row=r, start_column=1, end_row=r, end_column=4)
+        ws.row_dimensions[r].height = 100
+        r += 2
+        
+    industry_trends = data.get("industry_trends", "")
+    if industry_trends:
+        _sec(ws, r, "Industry Trends", 2); r += 1
+        c = ws.cell(row=r, column=1, value=industry_trends)
+        c.font = LBL_FONT; c.alignment = Alignment(wrap_text=True, vertical="top")
+        ws.merge_cells(start_row=r, start_column=1, end_row=r, end_column=4)
+        ws.row_dimensions[r].height = 120
+        r += 2
+        
+    _widths(ws, [100])
+
+
 def _sources_sheet(wb, data):
     ws = wb.create_sheet("Sources")
     _title(ws, "Sources", "All data sources and retrieval dates")
@@ -426,9 +558,11 @@ def build_workbook(data: dict, output_path: str) -> None:
     _cashflow_sheet(wb, data)
     _ratios_sheet(wb, data)
     _dupont_sheet(wb, data)
+    _dcf_sheet(wb, data)
     _growth_sheet(wb, data)
     _peers_sheet(wb, data)
     _swot_sheet(wb, data)
+    _narrative_sheet(wb, data)
     _sources_sheet(wb, data)
     wb.save(output_path)
 

builders/build_pptx.py

@@ -226,17 +226,17 @@ def build_presentation(data: Dict, output_path: str):
             _style_table_cell(table.cell(0, idx+1), str(y), FONT_MAJOR, 12, True, COLOR_LIGHT_BG, COLOR_DARK_SLATE)
             
         metrics = [
-            ("Revenue", "income_statement", 1.0),
-            ("Gross Profit", "income_statement", 1.0),
-            ("Operating Income (EBIT)", "income_statement", 1.0),
-            ("Net Income", "income_statement", 1.0),
-            ("Operating Cash Flow (OCF)", "cash_flow", 1.0)
+            ("Revenue", "revenue", "income_statement", 1.0),
+            ("Gross Profit", "gross_profit", "income_statement", 1.0),
+            ("Operating Income (EBIT)", "operating_income", "income_statement", 1.0),
+            ("Net Income", "net_income", "income_statement", 1.0),
+            ("Operating Cash Flow (OCF)", "cash_from_operations", "cash_flow", 1.0)
         ]
         
         currency = data.get("company", {}).get("currency", "USD")
         units = data.get("company", {}).get("units", "millions")
         
-        for r_idx, (m_label, sheet_name, scale) in enumerate(metrics):
+        for r_idx, (m_label, dict_key, sheet_name, scale) in enumerate(metrics):
             curr_row = r_idx + 1
             bg_color = COLOR_LIGHT_GRAY if curr_row % 2 == 1 else COLOR_LIGHT_BG
             
@@ -245,7 +245,7 @@ def build_presentation(data: Dict, output_path: str):
             
             # Values cells
             sheet_data = data.get(sheet_name, {})
-            val_list = sheet_data.get(m_label, [])
+            val_list = sheet_data.get(dict_key, [])
             for c_idx, y in enumerate(years):
                 val_str = "N/A"
                 if c_idx < len(val_list) and val_list[c_idx] is not None:

pipeline/analyst.py

@@ -183,6 +183,12 @@ def _build_context(data: Dict, meta: Dict) -> str:
                 f"ROE={_pct(p.get('roe'))} | "
                 f"D/E={_num(p.get('debt_to_equity'))}"
             )
+            
+    # Optional Annual Report Text (MD&A)
+    ar_text = data.get("annual_report_context", "")
+    if ar_text:
+        lines += ["", "=== ANNUAL REPORT EXCERPT (10-K MD&A) ==="]
+        lines.append(ar_text[:15000]) # Cap at 15k chars to save tokens
 
     return "\n".join(lines)
 
@@ -512,9 +518,9 @@ def _hf_generate(context: str, section: str, log_fn=None) -> Optional[str]:
 
 # ── Parse structured sections ─────────────────────────────────────────────────
 
-def _parse_swot(text: str) -> Dict[str, List[str]]:
+def _parse_swot(text: str) -> Dict:
     """Parse SWOT text into a structured dict."""
-    result = {"strengths": [], "weaknesses": [], "opportunities": [], "threats": []}
+    result = {"strengths": [], "weaknesses": [], "opportunities": [], "threats": [], "raw": text}
     current = None
     for line in text.split("\n"):
         l = line.strip()
@@ -527,7 +533,9 @@ def _parse_swot(text: str) -> Dict[str, List[str]]:
         elif l.upper().startswith("THREAT"):
             current = "threats"
         elif current and l and not l.upper() in ("STRENGTHS:", "WEAKNESSES:", "OPPORTUNITIES:", "THREATS:"):
-            result[current].append(l.lstrip("-• "))
+            clean = l.lstrip("-•* ")
+            if clean:
+                result[current].append(clean)
     return result
 
 

pipeline/fetcher.py

@@ -596,7 +596,6 @@ def fetch_peer_snapshot(ticker: str) -> Optional[Dict]:
     except Exception:
         return None
 
-
 def fetch_valuation(ticker: str) -> Dict:
     try:
         import yfinance as yf
@@ -613,8 +612,70 @@ def fetch_valuation(ticker: str) -> Dict:
     except Exception:
         return {}
 
+def _scrape_edgar_10k(ticker: str, log_fn=None) -> str:
+    """Fetch the latest 10-K for a US company from SEC EDGAR and extract a chunk of text."""
+    headers = {"User-Agent": "FinancialAnalyzer/1.0 (contact@example.com)"}
+    try:
+        if log_fn: log_fn(f"  → Looking up SEC CIK for {ticker}...")
+        r = requests.get("https://www.sec.gov/files/company_tickers.json", headers=headers, timeout=10)
+        tickers = r.json()
+        cik = None
+        for k, v in tickers.items():
+            if v["ticker"].upper() == ticker.upper():
+                cik = str(v["cik_str"]).zfill(10)
+                break
+        
+        if not cik:
+            if log_fn: log_fn("  ✗ CIK not found in SEC database.")
+            return ""
+            
+        if log_fn: log_fn(f"  → Found CIK {cik}. Fetching recent filings...")
+        sub_r = requests.get(f"https://data.sec.gov/submissions/CIK{cik}.json", headers=headers, timeout=10)
+        data = sub_r.json()
+        filings = data.get("filings", {}).get("recent", {})
+        
+        form_types = filings.get("form", [])
+        accessions = filings.get("accessionNumber", [])
+        primary_docs = filings.get("primaryDocument", [])
+        
+        idx = -1
+        for i, form in enumerate(form_types):
+            if form == "10-K":
+                idx = i
+                break
+                
+        if idx == -1:
+            if log_fn: log_fn("  ✗ No 10-K found in recent filings.")
+            return ""
+            
+        acc_num = accessions[idx]
+        acc_num_no_dash = acc_num.replace("-", "")
+        doc = primary_docs[idx]
+        
+        url = f"https://www.sec.gov/Archives/edgar/data/{int(cik)}/{acc_num_no_dash}/{doc}"
+        if log_fn: log_fn(f"  → Downloading 10-K from {url}...")
+        
+        doc_r = requests.get(url, headers=headers, timeout=20)
+        from bs4 import BeautifulSoup
+        soup = BeautifulSoup(doc_r.content, "lxml")
+        
+        text = soup.get_text(separator="\n", strip=True)
+        import re
+        match = re.search(r'(?i)Item\s+7\.\s*Management[’\']s\s+Discussion', text)
+        if match:
+            start = match.start()
+            extract = text[start:start+40000] 
+        else:
+            extract = text[len(text)//4 : len(text)//4 + 40000]
+            
+        if log_fn: log_fn("  ✓ 10-K text extracted successfully.")
+        return extract
+    except Exception as e:
+        if log_fn: log_fn(f"  ✗ Failed to scrape EDGAR: {str(e)[:80]}")
+        return ""
+
 
-# ── Main entry ────────────────────────────────────────────────────────────────
+# ── Main Fetcher ──────────────────────────────────────────────────────────────
 
 def fetch_all(meta: CompanyMeta, years: int = DEFAULT_YEARS,
               force_refresh: bool = False, log_fn=None) -> Dict:
@@ -684,6 +745,13 @@ def fetch_all(meta: CompanyMeta, years: int = DEFAULT_YEARS,
         return {"warnings": [msg], "years": [], "income_statement": {},
                 "balance_sheet": {}, "cash_flow": {}}
 
+    # 4. Optional Annual Report Scrape (for context)
+    if meta.currency != "INR":
+        ar_context = _scrape_edgar_10k(ticker, log_fn)
+        if ar_context:
+            result["annual_report_context"] = ar_context
+            warnings.append("AI analysis enhanced with raw MD&A excerpts from SEC EDGAR 10-K.")
+
     result["warnings"] = warnings
     result.setdefault("currency", meta.currency)
     set_cache(ticker, cache_key, result)