Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
|
@@ -94,9 +94,31 @@ def auto_detect(file_path):
|
|
| 94 |
m = re.match(r"spm_(\d{4})_(\w+)\.pdf", fname, re.IGNORECASE)
|
| 95 |
if m:
|
| 96 |
year, subject = m.groups()
|
| 97 |
-
return year, subject
|
| 98 |
return None, None
|
| 99 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
def process_pdf(file, subject, year):
|
| 101 |
raw = ocr_agent.extract_from_pdf(file)
|
| 102 |
cleaned = ocr_agent.clean_text(raw)
|
|
@@ -115,7 +137,8 @@ def prefill_subject_year(file):
|
|
| 115 |
if not file:
|
| 116 |
return "BM", "2018"
|
| 117 |
year, subject = auto_detect(file)
|
| 118 |
-
|
|
|
|
| 119 |
|
| 120 |
# ----------------- Gradio UI -----------------
|
| 121 |
with gr.Blocks() as demo:
|
|
@@ -189,3 +212,4 @@ if __name__ == "__main__":
|
|
| 189 |
|
| 190 |
|
| 191 |
|
|
|
|
|
|
| 94 |
m = re.match(r"spm_(\d{4})_(\w+)\.pdf", fname, re.IGNORECASE)
|
| 95 |
if m:
|
| 96 |
year, subject = m.groups()
|
| 97 |
+
return year, subject
|
| 98 |
return None, None
|
| 99 |
|
| 100 |
+
def normalize_subject(subject):
|
| 101 |
+
"""Normalize subject names to match dropdown choices."""
|
| 102 |
+
valid_subjects = ["BM", "English", "Math", "History", "Science", "MoralStudies"]
|
| 103 |
+
if not subject:
|
| 104 |
+
return "BM"
|
| 105 |
+
subject = subject.strip().lower()
|
| 106 |
+
mapping = {
|
| 107 |
+
"bm": "BM",
|
| 108 |
+
"bahasa": "BM",
|
| 109 |
+
"bahasamelayu": "BM",
|
| 110 |
+
"b.m": "BM",
|
| 111 |
+
"english": "English",
|
| 112 |
+
"math": "Math",
|
| 113 |
+
"mathematics": "Math",
|
| 114 |
+
"history": "History",
|
| 115 |
+
"sejarah": "History",
|
| 116 |
+
"science": "Science",
|
| 117 |
+
"moral": "MoralStudies",
|
| 118 |
+
"moralstudies": "MoralStudies",
|
| 119 |
+
}
|
| 120 |
+
return mapping.get(subject, "BM") if subject in mapping else subject.capitalize()
|
| 121 |
+
|
| 122 |
def process_pdf(file, subject, year):
|
| 123 |
raw = ocr_agent.extract_from_pdf(file)
|
| 124 |
cleaned = ocr_agent.clean_text(raw)
|
|
|
|
| 137 |
if not file:
|
| 138 |
return "BM", "2018"
|
| 139 |
year, subject = auto_detect(file)
|
| 140 |
+
subject = normalize_subject(subject)
|
| 141 |
+
return subject, year if year else "2018"
|
| 142 |
|
| 143 |
# ----------------- Gradio UI -----------------
|
| 144 |
with gr.Blocks() as demo:
|
|
|
|
| 212 |
|
| 213 |
|
| 214 |
|
| 215 |
+
|