Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -85,6 +85,49 @@ def material_from(text: str) -> str:
|
|
| 85 |
m = re.search(r"(SERIOPLAST.*?RESIN)", text, re.I)
|
| 86 |
return m.group(1).strip() if m else ""
|
| 87 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
# --------------------- AGGIUNTA RICHIESTA: PIECE da "Packaging Component Type" ---------------------
|
| 89 |
_ALLOWED_PIECES = {
|
| 90 |
"ribbon": "ribbon",
|
|
@@ -239,13 +282,16 @@ def parse_record(pages: List[str], source_name: str) -> Dict[str, str]:
|
|
| 239 |
gcas = material_ref_gcas_from(full)
|
| 240 |
mfam = material_family_from(full)
|
| 241 |
|
|
|
|
|
|
|
|
|
|
| 242 |
return {
|
| 243 |
"Piece": piece or "",
|
| 244 |
"SKU": sku or "",
|
| 245 |
"Title": title or "",
|
| 246 |
"Capacity": cap or "",
|
| 247 |
"% Recycled": "–",
|
| 248 |
-
"Weight": "–",
|
| 249 |
"Color": color or "",
|
| 250 |
"Material / Resin": material or "",
|
| 251 |
"Class": cls or "",
|
|
@@ -313,5 +359,5 @@ with c2:
|
|
| 313 |
bio = io.BytesIO()
|
| 314 |
with pd.ExcelWriter(bio, engine="openpyxl") as xw:
|
| 315 |
df.to_excel(xw, index=False, sheet_name="data")
|
| 316 |
-
st.download_button("⬇️ Excel", bio.getvalue(), "table.xlsx", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
|
| 317 |
|
|
|
|
| 85 |
m = re.search(r"(SERIOPLAST.*?RESIN)", text, re.I)
|
| 86 |
return m.group(1).strip() if m else ""
|
| 87 |
|
| 88 |
+
# --- (AGGIUNTA) WEIGHT PARSER ----------------------------------------------
|
| 89 |
+
WEIGHT_TOL_RE = re.compile(
|
| 90 |
+
r"\bWeight\b[^\n\r]{0,15}?([0-9]+(?:[.,][0-9]+)?)\s*(?:±|\+/?-|\+-)\s*([0-9]+(?:[.,][0-9]+)?)\s*(mg|g|kg)?",
|
| 91 |
+
re.I,
|
| 92 |
+
)
|
| 93 |
+
WEIGHT_SIMPLE_RE = re.compile(
|
| 94 |
+
r"\bWeight\b[^\n\r]{0,15}?([0-9]+(?:[.,][0-9]+)?)\s*(mg|g|kg)\b",
|
| 95 |
+
re.I,
|
| 96 |
+
)
|
| 97 |
+
WEIGHT_INLINE_RE = re.compile(
|
| 98 |
+
r"\b([0-9]+(?:[.,][0-9]+)?)\s*(?:±|\+/?-|\+-)\s*([0-9]+(?:[.,][0-9]+)?)\s*(mg|g|kg)\b",
|
| 99 |
+
re.I,
|
| 100 |
+
)
|
| 101 |
+
|
| 102 |
+
def _norm_num(s: str) -> str:
|
| 103 |
+
return (s or "").replace(",", ".").strip().rstrip(".")
|
| 104 |
+
|
| 105 |
+
def weight_from(text: str) -> str:
|
| 106 |
+
# 1) match con tolleranza
|
| 107 |
+
m = WEIGHT_TOL_RE.search(text or "")
|
| 108 |
+
if m:
|
| 109 |
+
val = _norm_num(m.group(1))
|
| 110 |
+
tol = _norm_num(m.group(2))
|
| 111 |
+
unit = (m.group(3) or "g").lower()
|
| 112 |
+
return f"{val} ± {tol} {unit}"
|
| 113 |
+
# 2) match semplice con unità
|
| 114 |
+
m = WEIGHT_SIMPLE_RE.search(text or "")
|
| 115 |
+
if m:
|
| 116 |
+
val = _norm_num(m.group(1))
|
| 117 |
+
unit = (m.group(2) or "g").lower()
|
| 118 |
+
return f"{val} {unit}"
|
| 119 |
+
# 3) riga per riga per casi OCR
|
| 120 |
+
for line in (text or "").splitlines():
|
| 121 |
+
if "weight" in line.lower():
|
| 122 |
+
m2 = WEIGHT_INLINE_RE.search(line)
|
| 123 |
+
if m2:
|
| 124 |
+
val = _norm_num(m2.group(1))
|
| 125 |
+
tol = _norm_num(m2.group(2))
|
| 126 |
+
unit = (m2.group(3) or "g").lower()
|
| 127 |
+
return f"{val} ± {tol} {unit}"
|
| 128 |
+
return ""
|
| 129 |
+
# ---------------------------------------------------------------------------
|
| 130 |
+
|
| 131 |
# --------------------- AGGIUNTA RICHIESTA: PIECE da "Packaging Component Type" ---------------------
|
| 132 |
_ALLOWED_PIECES = {
|
| 133 |
"ribbon": "ribbon",
|
|
|
|
| 282 |
gcas = material_ref_gcas_from(full)
|
| 283 |
mfam = material_family_from(full)
|
| 284 |
|
| 285 |
+
# (AGGIUNTA) estrai peso
|
| 286 |
+
wght = weight_from(full)
|
| 287 |
+
|
| 288 |
return {
|
| 289 |
"Piece": piece or "",
|
| 290 |
"SKU": sku or "",
|
| 291 |
"Title": title or "",
|
| 292 |
"Capacity": cap or "",
|
| 293 |
"% Recycled": "–",
|
| 294 |
+
"Weight": wght or "–",
|
| 295 |
"Color": color or "",
|
| 296 |
"Material / Resin": material or "",
|
| 297 |
"Class": cls or "",
|
|
|
|
| 359 |
bio = io.BytesIO()
|
| 360 |
with pd.ExcelWriter(bio, engine="openpyxl") as xw:
|
| 361 |
df.to_excel(xw, index=False, sheet_name="data")
|
| 362 |
+
st.download_button("⬇️ Excel", bio.getvalue(), "table.xlsx", "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
|
| 363 |
|