Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,268 +1,146 @@
 import re
-from io import BytesIO
-from datetime import date
+from datetime import datetime
+from typing import List
 
-import streamlit as st
-from docx import Document
-from docx.shared import Inches, Pt
-from docx.enum.text import WD_ALIGN_PARAGRAPH
-from docx.oxml import OxmlElement
-from docx.oxml.ns import qn
+import gradio as gr
+from pypdf import PdfReader
 
-# -----------------------------
-# Config
-# -----------------------------
-st.set_page_config(page_title="Checklist de Remessa", layout="centered")
+from reportlab.lib.pagesizes import A4
+from reportlab.pdfgen import canvas
+from reportlab.lib.units import mm
 
-CLIENTE_FIXO = "ArtStones"
-CHAVE_RE = re.compile(r"\b(\d{44})\b")
 
+KEY_RE = re.compile(r"\b\d{44}\b")
 
-def extrair_chaves(texto: str) -> list[str]:
-    if not texto:
-        return []
-    return CHAVE_RE.findall(texto)
 
-
-def listar_duplicadas(lista: list[str]) -> list[str]:
+def extract_keys_from_pdf(pdf_path: str) -> List[str]:
+    reader = PdfReader(pdf_path)
+    found: List[str] = []
     seen = set()
-    dups = []
-    for x in lista:
-        if x in seen and x not in dups:
-            dups.append(x)
-        seen.add(x)
-    return dups
-
-
-def chunk_3cols(items: list[str]) -> tuple[list[str], list[str], list[str]]:
-    """Divide em 3 colunas o mais equilibrado possível."""
-    n = len(items)
-    if n == 0:
-        return [], [], []
-    a = (n + 2) // 3  # ceil(n/3)
-    b = (n + 1) // 3
-    col1 = items[:a]
-    col2 = items[a : a + b]
-    col3 = items[a + b :]
-    return col1, col2, col3
-
-
-def set_cell(cell, title: str, value: str = "", font_size=11, bold_title=True, align="center"):
-    """
-    Escreve título + valor na célula e centraliza (bonitinho).
-    """
-    cell.text = ""
-    p = cell.paragraphs[0]
-    p.alignment = {
-        "left": WD_ALIGN_PARAGRAPH.LEFT,
-        "center": WD_ALIGN_PARAGRAPH.CENTER,
-        "right": WD_ALIGN_PARAGRAPH.RIGHT,
-    }[align]
-
-    r1 = p.add_run(title + ("\n" if value else ""))
-    r1.bold = bold_title
-    r1.font.size = Pt(font_size)
-    r1.font.name = "Calibri"
-    r1._element.rPr.rFonts.set(qn("w:eastAsia"), "Calibri")
-
-    if value:
-        r2 = p.add_run(value)
-        r2.bold = False
-        r2.font.size = Pt(font_size)
-        r2.font.name = "Calibri"
-        r2._element.rPr.rFonts.set(qn("w:eastAsia"), "Calibri")
-
-
-def set_cell_lines(cell, header: str, lines: list[str], font_size=9, align="center"):
-    """
-    Cabeçalho + lista de chaves em linhas. Alinhamento central para ficar como o modelo.
-    """
-    cell.text = ""
-    p = cell.paragraphs[0]
-    p.alignment = {
-        "left": WD_ALIGN_PARAGRAPH.LEFT,
-        "center": WD_ALIGN_PARAGRAPH.CENTER,
-        "right": WD_ALIGN_PARAGRAPH.RIGHT,
-    }[align]
-
-    rh = p.add_run(header + ("\n" if lines else ""))
-    rh.bold = True
-    rh.font.size = Pt(10)
-    rh.font.name = "Calibri"
-    rh._element.rPr.rFonts.set(qn("w:eastAsia"), "Calibri")
-
-    if lines:
-        rl = p.add_run("\n".join(lines))
-        rl.bold = False
-        rl.font.size = Pt(font_size)
-        rl.font.name = "Calibri"
-        rl._element.rPr.rFonts.set(qn("w:eastAsia"), "Calibri")
-
-
-def remove_table_borders(table):
-    tbl = table._tbl
-    tblPr = tbl.tblPr
-    borders = OxmlElement("w:tblBorders")
-    for edge in ("top", "left", "bottom", "right", "insideH", "insideV"):
-        elem = OxmlElement(f"w:{edge}")
-        elem.set(qn("w:val"), "nil")
-        borders.append(elem)
-    tblPr.append(borders)
-
-
-def build_docx_layout_modelo(cliente: str, data_coleta: str, hora: str, eco: list[str], rap: list[str]) -> bytes:
-    doc = Document()
-
-    # Margens
-    section = doc.sections[0]
-    section.top_margin = Inches(0.55)
-    section.bottom_margin = Inches(0.55)
-    section.left_margin = Inches(0.55)
-    section.right_margin = Inches(0.55)
-
-    # Tabela principal (igual ao modelo)
-    table = doc.add_table(rows=4, cols=5)
-    table.style = "Table Grid"
-
-    # --- Linha 0 (cabeçalho) ---
-    # CLIENTE (col 0), DATA (mescla col 1-3), HORA (col 4)
-    cell_cliente = table.cell(0, 0)
-    cell_data = table.cell(0, 1).merge(table.cell(0, 2)).merge(table.cell(0, 3))
-    cell_hora = table.cell(0, 4)
-
-    set_cell(cell_cliente, "CLIENTE:", cliente, font_size=11, bold_title=True, align="center")
-    set_cell(cell_data, "DATA DA COLETA:", data_coleta, font_size=11, bold_title=True, align="center")
-    set_cell(cell_hora, "HORA DA COLETA:", hora if hora else "_____ : _____", font_size=11, bold_title=True, align="center")
-
-    # --- Linha 1 (ECONÔMICO) ---
-    eco_left = table.cell(1, 0).merge(table.cell(1, 1))
-    eco_mid = table.cell(1, 2)
-    eco_right = table.cell(1, 3).merge(table.cell(1, 4))
-
-    e1, e2, e3 = chunk_3cols(eco)
-    set_cell_lines(eco_left, f"ECONÔMICO = {len(eco)}", e1, font_size=9, align="center")
-    set_cell_lines(eco_mid, "", e2, font_size=9, align="center")
-    set_cell_lines(eco_right, "", e3, font_size=9, align="center")
-
-    # --- Linha 2 (RÁPIDO) ---
-    rap_left = table.cell(2, 0).merge(table.cell(2, 1))
-    rap_mid = table.cell(2, 2)
-    rap_right = table.cell(2, 3).merge(table.cell(2, 4))
-
-    r1, r2, r3 = chunk_3cols(rap)
-    set_cell_lines(rap_left, f"RÁPIDO = {len(rap)}", r1, font_size=9, align="center")
-    set_cell_lines(rap_mid, "", r2, font_size=9, align="center")
-    set_cell_lines(rap_right, "", r3, font_size=9, align="center")
-
-    # --- Linha 3 (TOTAL) ---
-    total_cell = table.cell(3, 0)
-    for c in range(1, 5):
-        total_cell = total_cell.merge(table.cell(3, c))
-    set_cell(total_cell, "TOTAL DA REMESSA:", f"{len(eco) + len(rap)} VOLUMES", font_size=11, bold_title=True, align="center")
-
-    # Espaço + assinaturas (centralizadas)
-    doc.add_paragraph("\n")
-
-    sign = doc.add_table(rows=2, cols=2)
-    remove_table_borders(sign)
-
-    set_cell(sign.cell(0, 0), "", "______________________________", font_size=11, bold_title=False, align="center")
-    set_cell(sign.cell(0, 1), "", "______________________________", font_size=11, bold_title=False, align="center")
-    set_cell(sign.cell(1, 0), "ASSINATURA DO REPRESENTANTE", "", font_size=10, bold_title=True, align="center")
-    set_cell(sign.cell(1, 1), "ASSINATURA DO MOTORISTA", "", font_size=10, bold_title=True, align="center")
-
-    bio = BytesIO()
-    doc.save(bio)
-    return bio.getvalue()
-
-
-# -----------------------------
-# Estado
-# -----------------------------
-if "eco" not in st.session_state:
-    st.session_state.eco = []
-if "rap" not in st.session_state:
-    st.session_state.rap = []
-
-# -----------------------------
-# UI
-# -----------------------------
-st.title("📋📦 Checklist de Remessa (Econômico + Rápido)")
-st.caption(f"Cliente: {CLIENTE_FIXO}")
-
-data_coleta = st.text_input("Data da coleta:", value=date.today().strftime("%d/%m/%Y"))
-hora = st.text_input("Hora da coleta:", value="_____ : _____")
-
-modo = st.radio("Tipo de frete atual:", ["ECONÔMICO", "RÁPIDO"], horizontal=True)
-
-
-def adicionar_scan():
-    raw = st.session_state.scan_input.strip()
-    st.session_state.scan_input = ""
-
-    chaves = extrair_chaves(raw)
-    if not chaves:
-        st.warning("Nenhuma chave (44 dígitos) detectada nesse scan.")
-        return
-
-    destino = st.session_state.eco if modo == "ECONÔMICO" else st.session_state.rap
-    destino.extend(chaves)
-
-
-st.text_input(
-    "Bipe aqui (o scanner já quebra linha):",
-    key="scan_input",
-    on_change=adicionar_scan,
-    placeholder="Bipe/cole a chave de 44 dígitos…",
-)
-
-eco = st.session_state.eco
-rap = st.session_state.rap
-total = len(eco) + len(rap)
-
-c1, c2, c3 = st.columns(3)
-c1.success(f"🟩 Econ.: {len(eco)}")
-c2.info(f"🟦 Ráp.: {len(rap)}")
-c3.warning(f"🚚 Total: {total}")
-
-dups_eco = listar_duplicadas(eco)
-dups_rap = listar_duplicadas(rap)
-if dups_eco or dups_rap:
-    with st.expander("⚠️ Ver duplicadas"):
-        if dups_eco:
-            st.write(f"Econômico ({len(dups_eco)}):")
-            st.code("\n".join(dups_eco[:200]))
-        if dups_rap:
-            st.write(f"Rápido ({len(dups_rap)}):")
-            st.code("\n".join(dups_rap[:200]))
-
-st.divider()
-
-doc_bytes = build_docx_layout_modelo(CLIENTE_FIXO, data_coleta, hora, eco, rap)
-
-# Nome padronizado: "Xml Mandae - DD.MM.AAAA.docx"
-data_nome = data_coleta.replace("/", ".")
-nome_arquivo = f"Xml Mandae - {data_nome}.docx"
-
-st.download_button(
-    "⬇️ Baixar checklist (.docx) para assinatura",
-    data=doc_bytes,
-    file_name=nome_arquivo,
-    mime="application/vnd.openxmlformats-officedocument.wordprocessingml.document",
-)
-
-colR1, colR2 = st.columns(2)
-with colR1:
-    if st.button("🧹 Zerar tudo"):
-        st.session_state.eco = []
-        st.session_state.rap = []
-        st.rerun()
-
-with colR2:
-    if st.button("↩️ Remover última chave do modo atual"):
-        if modo == "ECONÔMICO" and st.session_state.eco:
-            st.session_state.eco.pop()
-            st.rerun()
-        if modo == "RÁPIDO" and st.session_state.rap:
-            st.session_state.rap.pop()
-            st.rerun()
+
+    for page in reader.pages:
+        text = page.extract_text() or ""
+        for k in KEY_RE.findall(text):
+            if k not in seen:
+                seen.add(k)
+                found.append(k)
+    return found
+
+
+def safe_date_default() -> str:
+    return datetime.now().strftime("%d/%m/%Y")
+
+
+def render_print_pdf(out_path: str, data_coleta: str, hora_coleta: str, keys: List[str]):
+    c = canvas.Canvas(out_path, pagesize=A4)
+    w, h = A4
+
+    left = 20 * mm
+    top = h - 20 * mm
+    line_h = 6.2 * mm
+
+    def draw(text: str, y: float, bold=False, size=11) -> float:
+        c.setFont("Helvetica-Bold" if bold else "Helvetica", size)
+        c.drawString(left, y, text)
+        return y - line_h
+
+    y = top
+
+    # Cabeçalho
+    y = draw("CLIENTE:", y)
+    y -= line_h * 1.3
+
+    y = draw("DATA DA COLETA:", y)
+    y = draw(data_coleta, y)
+    y -= line_h * 0.6
+
+    y = draw("HORA DA COLETA:", y)
+    y = draw("_____ : _____", y)
+    y -= line_h * 0.9
+
+    # ===== ALTERAÇÃO AQUI =====
+    y = draw("CHAVES DE ACESSO:", y)
+    y -= line_h * 0.2
+    # =========================
+
+    c.setFont("Helvetica", 10.8)
+    for k in keys:
+        if y < 40 * mm:
+            c.showPage()
+            y = top
+            c.setFont("Helvetica", 10.8)
+        c.drawString(left, y, k)
+        y -= line_h * 0.85
+
+    y -= line_h * 0.8
+    y = draw(f"TOTAL DA REMESSA:  {len(keys)} VOLUMES", y)
+
+    # Assinaturas
+    if y < 55 * mm:
+        c.showPage()
+
+    y_sig = 25 * mm
+    c.setFont("Helvetica", 9.5)
+    c.drawString(left + 70 * mm, y_sig + 12, "ASSINATURA DO REPRESENTANTE")
+    c.drawString(left + 140 * mm, y_sig + 12, "ASSINATURA DO MOTORISTA")
+    c.line(left + 55 * mm, y_sig + 10, left + 118 * mm, y_sig + 10)
+    c.line(left + 130 * mm, y_sig + 10, left + 193 * mm, y_sig + 10)
+
+    c.save()
+
+
+def build_txt(out_path: str, keys: List[str]):
+    with open(out_path, "w", encoding="utf-8") as f:
+        f.write("\n".join(keys))
+        f.write("\n")
+
+
+def run(pdf_file, data_coleta, hora_coleta):
+    if pdf_file is None:
+        raise gr.Error("Envie um PDF.")
+
+    keys = extract_keys_from_pdf(pdf_file)
+    if not keys:
+        raise gr.Error("Não encontrei chaves de 44 dígitos no PDF.")
+
+    ts = datetime.now().strftime("%Y%m%d-%H%M%S")
+    out_pdf = f"chaves_prontas_{ts}.pdf"
+    out_txt = f"chaves_{ts}.txt"
+
+    render_print_pdf(
+        out_pdf,
+        data_coleta=data_coleta or safe_date_default(),
+        hora_coleta=hora_coleta or "_____ : _____",
+        keys=keys,
+    )
+    build_txt(out_txt, keys)
+
+    preview = (
+        f"Total extraído: {len(keys)}\n\n"
+        "Primeiras chaves:\n" + "\n".join(keys[:10])
+    )
+    return preview, out_pdf, out_txt
+
+
+with gr.Blocks(title="Extrator de Chaves NF-e") as demo:
+    gr.Markdown("### Envie um PDF e gere um arquivo pronto para imprimir.")
+
+    with gr.Row():
+        pdf = gr.File(label="PDF", file_types=[".pdf"])
+        preview = gr.Textbox(label="Prévia", lines=12)
+
+    with gr.Row():
+        data_coleta = gr.Textbox(label="Data da coleta", value=safe_date_default())
+        hora_coleta = gr.Textbox(label="Hora da coleta", value="_____ : _____")
+
+    btn = gr.Button("Gerar arquivos")
+    out_pdf = gr.File(label="PDF pronto para imprimir")
+    out_txt = gr.File(label="TXT (opcional)")
+
+    btn.click(
+        fn=run,
+        inputs=[pdf, data_coleta, hora_coleta],
+        outputs=[preview, out_pdf, out_txt],
+    )
+
+demo.launch()