File size: 10,941 Bytes
9c6961c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
import gradio as gr
import json
import os
import random
import glob
from datetime import datetime

# --- Configuration & Folder Setup ---
DATA_PATH = '/home/mshahidul/readctrl/data/extracting_subclaim/extracted_subclaims_syn_data_with_gs_summary_en.json'
KEY_DATA_PATH = '/home/mshahidul/readctrl/data/key_subclaims_testing/key_subclaims.json'
BASE_SAVE_DIR = '/home/mshahidul/readctrl/data/thresold_finding/'

# --- Data Loading ---
with open(DATA_PATH, 'r') as f:
    data = json.load(f)

NUM_SAMPLES = 10
random.seed(42) 
all_possible_indices = list(range(len(data)))
shuffled_indices = random.sample(all_possible_indices, min(NUM_SAMPLES, len(data)))

with open(KEY_DATA_PATH, 'r') as f:
    key_data = json.load(f)
key_lookup = {item['index']: item['llm_output'] for item in key_data}

# --- Helper Functions ---
def get_user_dir(username):
    if not username: return None
    safe_name = "".join(x for x in username if x.isalnum()).lower()
    user_path = os.path.join(BASE_SAVE_DIR, safe_name)
    os.makedirs(user_path, exist_ok=True)
    return user_path

def get_last_progress(username):
    user_dir = get_user_dir(username)
    files = glob.glob(os.path.join(user_dir, "seq*_*.json"))
    if not files: return 0
    indices = []
    for f in files:
        try:
            indices.append(int(os.path.basename(f).split('_')[0].replace('seq', '')))
        except: continue
    return min(max(indices) + 1, NUM_SAMPLES - 1) if indices else 0

# --- Core Logic ---
def load_example(progress_index, username):
    if not username:
        return [gr.update(value="### ⚠️ Please enter your name and click Login")] + [gr.skip()]*10

    if progress_index >= len(shuffled_indices):
        return ["### πŸŽ‰ All Samples Complete!", "Done", [], "0%", "0%", "0%", gr.update(choices=[], value=[]), gr.update(choices=[], value=[]), gr.update(choices=[], value=[]), "Session Finished", progress_index]

    actual_data_index = shuffled_indices[progress_index]
    record = data[actual_data_index]
    
    random.seed(actual_data_index)
    source_type = random.choice(["Full Original Text", "Gold Summary"])
    text_content, subclaims = (record['fulltext'], record['fulltext_subclaims']) if source_type == "Full Original Text" else (record['summary'], record['summary_subclaims'])
        
    user_dir = get_user_dir(username)
    existing_files = glob.glob(os.path.join(user_dir, f"seq{progress_index}_*.json"))
    
    if existing_files:
        with open(existing_files[0], 'r') as f:
            saved = json.load(f)
            low_val = saved['annotations']['low']['subclaims']
            int_val = saved['annotations']['intermediate']['subclaims']
            prof_val = saved['annotations']['proficient']['subclaims']
        status_msg = f"πŸ“‚ [Sequence {progress_index}] Previously saved data loaded."
    else:
        key_items = key_lookup.get(actual_data_index, {}).get('key_source_text_subclaims' if source_type == "Full Original Text" else 'key_gold_summary_subclaims', [])
        indices = []
        for item in key_items:
            try: indices.append(int(item.get("source_subclaim_id" if source_type == "Full Original Text" else "gold_subclaim_id", "").split('-')[-1]))
            except: continue
        default_sel = [subclaims[i] for i in indices if 0 <= i < len(subclaims)]
        low_val, int_val, prof_val = default_sel, default_sel, default_sel
        status_msg = f"πŸ†• [Sequence {progress_index}] New record loaded."

    source_info = f"### Instance: {progress_index + 1}/{len(shuffled_indices)} | User: **{username}** | Source: **{source_type}**"
    
    # Calculate initial percentages for UI
    total = len(subclaims) if subclaims else 1
    l_p, i_p, p_p = f"{(len(low_val)/total*100):.1f}%", f"{(len(int_val)/total*100):.1f}%", f"{(len(prof_val)/total*100):.1f}%"

    return [
        source_info, text_content, subclaims, l_p, i_p, p_p,
        gr.update(choices=subclaims, value=low_val), 
        gr.update(choices=subclaims, value=int_val), 
        gr.update(choices=subclaims, value=prof_val),
        status_msg, progress_index
    ]

def handle_save(username, progress_index, source_info, low_sel, int_sel, prof_sel, subclaims):
    if not username or username.strip() == "":
        gr.Warning("User name missing! Please enter name.")
        return "❌ Error: Username Required"
    
    if not (len(low_sel) <= len(int_sel) <= len(prof_sel)):
        gr.Warning("Hierarchy Error: Selections must follow Low βŠ† Intermediate βŠ† Proficient.")
        return "❌ Save Failed: Hierarchy Violation"

    user_dir = get_user_dir(username)
    actual_data_index = shuffled_indices[progress_index]
    stype = "Full Original Text" if "Full Original Text" in source_info else "Gold Summary"
    
    # Calculate Percentages for saving
    total_count = len(subclaims) if subclaims else 1
    low_pct_val = (len(low_sel) / total_count) * 100
    int_pct_val = (len(int_sel) / total_count) * 100
    prof_pct_val = (len(prof_sel) / total_count) * 100

    result = {
        "annotator": username,
        "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
        "progress_sequence": progress_index,
        "original_data_index": actual_data_index,
        "source_type": stype,
        "total_subclaims": total_count,
        "annotations": {
            "low": {
                "count": len(low_sel), 
                "percentage": f"{low_pct_val:.2f}%",
                "subclaims": low_sel
            },
            "intermediate": {
                "count": len(int_sel), 
                "percentage": f"{int_pct_val:.2f}%",
                "subclaims": int_sel
            },
            "proficient": {
                "count": len(prof_sel), 
                "percentage": f"{prof_pct_val:.2f}%",
                "subclaims": prof_sel
            }
        }
    }
    
    filename = f"seq{progress_index}_record{actual_data_index}.json"
    file_path = os.path.join(user_dir, filename)
    
    with open(file_path, 'w') as f:
        json.dump(result, f, indent=4)
    
    gr.Info(f"Record {progress_index + 1} saved successfully!")
    return f"βœ… Last saved: {datetime.now().strftime('%H:%M:%S')}"

def navigate(direction, current_idx):
    return max(0, min(current_idx + direction, NUM_SAMPLES - 1))

def sync_logic(low, inter, prof, total, trigger_type):
    if trigger_type == "low":
        inter, prof = list(set(inter) | set(low)), list(set(prof) | set(inter) | set(low))
    elif trigger_type == "inter":
        prof, low = list(set(prof) | set(inter)), list(set(low) & set(inter))
    else:
        inter, low = list(set(inter) & set(prof)), list(set(low) & set(inter) & set(prof))
    
    calc_pct = lambda x: f"{(len(x)/len(total)*100):.1f}%" if total else "0%"
    return calc_pct(low), calc_pct(inter), calc_pct(prof), gr.update(value=low), gr.update(value=inter), gr.update(value=prof)

# --- UI Definition ---
with gr.Blocks(theme=gr.themes.Soft(), title="Medical Literacy Tool") as demo:
    index_state = gr.State(0)
    subclaim_list_state = gr.State([])
    
    with gr.Row():
        with gr.Column(scale=2):
            user_input = gr.Textbox(label="Annotator Name", placeholder="e.g., Shahidul", interactive=True)
            load_btn = gr.Button("πŸš€ Login / Resume Session", variant="primary")
        with gr.Column(scale=3):
            with gr.Accordion("πŸ“– View Task Instructions", open=False):
                try:
                    with open("/home/mshahidul/readctrl/code/interface/instructions", "r") as f:
                        gr.Markdown(f.read())
                except:
                    gr.Markdown("### Instructions\n- Adjust subclaims for literacy levels.\n- **Saving:** Overwrites previous edits for the same record.")

    gr.HTML("<hr>")

    with gr.Row():
        with gr.Column(scale=1, variant="panel"):
            source_display = gr.Markdown("### Please login to begin.")
            progress_bar = gr.Slider(label="Progress", minimum=0, maximum=NUM_SAMPLES-1, step=1, interactive=False)
            text_viewer = gr.Textbox(label="Reference Text", interactive=False, lines=18)
            save_status = gr.Markdown("Status: Waiting for login...")

        with gr.Column(scale=2):
            with gr.Row():
                with gr.Column():
                    gr.Markdown("### 🟒 Low")
                    low_pct = gr.Label(label="Coverage", value="0%")
                    low_check = gr.CheckboxGroup(label="Subclaims", choices=[])
                with gr.Column():
                    gr.Markdown("### 🟑 Intermediate")
                    int_pct = gr.Label(label="Coverage", value="0%")
                    int_check = gr.CheckboxGroup(label="Subclaims", choices=[])
                with gr.Column():
                    gr.Markdown("### πŸ”΄ Proficient")
                    prof_pct = gr.Label(label="Coverage", value="0%")
                    prof_check = gr.CheckboxGroup(label="Subclaims", choices=[])

            with gr.Row():
                prev_btn = gr.Button("⬅️ Previous")
                save_btn = gr.Button("πŸ’Ύ Save Changes", variant="primary")
                next_btn = gr.Button("Next ➑️")

    # --- Event Handlers ---
    load_btn.click(lambda u: (get_last_progress(u), f"Session for {u} active."), [user_input], [index_state, save_status]).then(
        load_example, [index_state, user_input], 
        [source_display, text_viewer, subclaim_list_state, low_pct, int_pct, prof_pct, low_check, int_check, prof_check, save_status, progress_bar]
    )

    save_btn.click(handle_save, [user_input, index_state, source_display, low_check, int_check, prof_check, subclaim_list_state], [save_status])

    next_btn.click(navigate, [gr.Number(1, visible=False), index_state], [index_state]).then(
        load_example, [index_state, user_input], 
        [source_display, text_viewer, subclaim_list_state, low_pct, int_pct, prof_pct, low_check, int_check, prof_check, save_status, progress_bar]
    )
    
    prev_btn.click(navigate, [gr.Number(-1, visible=False), index_state], [index_state]).then(
        load_example, [index_state, user_input], 
        [source_display, text_viewer, subclaim_list_state, low_pct, int_pct, prof_pct, low_check, int_check, prof_check, save_status, progress_bar]
    )

    # Sync Logic for Hierarchy
    low_check.input(lambda l,i,p,t: sync_logic(l,i,p,t,"low"), [low_check, int_check, prof_check, subclaim_list_state], [low_pct, int_pct, prof_pct, low_check, int_check, prof_check])
    int_check.input(lambda l,i,p,t: sync_logic(l,i,p,t,"inter"), [low_check, int_check, prof_check, subclaim_list_state], [low_pct, int_pct, prof_pct, low_check, int_check, prof_check])
    prof_check.input(lambda l,i,p,t: sync_logic(l,i,p,t,"prof"), [low_check, int_check, prof_check, subclaim_list_state], [low_pct, int_pct, prof_pct, low_check, int_check, prof_check])

if __name__ == "__main__":
    demo.launch(share=True)