Update app.py via AI Editor
Browse files
app.py
CHANGED
@@ -12,49 +12,43 @@ import threading
|
|
12 |
import tempfile
|
13 |
import shutil
|
14 |
import logging
|
15 |
-
from flask import request
|
16 |
from dash.exceptions import PreventUpdate
|
17 |
import PyPDF2
|
18 |
import docx
|
19 |
import chardet
|
20 |
|
21 |
-
# --- Logging Setup ---
|
22 |
logging.basicConfig(level=logging.INFO)
|
23 |
logger = logging.getLogger("maiko_matrix_app")
|
24 |
|
25 |
-
# --- App and Session Setup ---
|
26 |
-
app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])
|
27 |
-
server = app.server
|
28 |
-
|
29 |
-
openai.api_key = os.environ.get('OPENAI_API_KEY')
|
30 |
-
|
31 |
-
# Session management globals
|
32 |
SESSION_DATA = {}
|
33 |
SESSION_LOCKS = {}
|
34 |
|
35 |
-
def get_session_id():
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
|
|
|
|
|
|
42 |
|
43 |
def get_session_data():
|
44 |
-
|
45 |
-
if
|
46 |
-
SESSION_DATA[
|
47 |
-
'uploaded_files': {},
|
48 |
-
'file_texts': {},
|
49 |
'current_matrix': None,
|
50 |
'matrix_type': None,
|
51 |
-
'temp_dir': tempfile.mkdtemp(prefix=f"maiko_{
|
52 |
}
|
53 |
-
SESSION_LOCKS[
|
54 |
-
return SESSION_DATA[
|
55 |
|
56 |
def cleanup_session_tempdirs():
|
57 |
-
# Called optionally on server shutdown for cleanup
|
58 |
for sess in SESSION_DATA.values():
|
59 |
try:
|
60 |
shutil.rmtree(sess['temp_dir'])
|
@@ -81,7 +75,102 @@ matrix_types = {
|
|
81 |
"SWOT Matrix": "Create a matrix analyzing Strengths, Weaknesses, Opportunities, and Threats."
|
82 |
}
|
83 |
|
84 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
85 |
app.layout = dbc.Container([
|
86 |
dbc.Row([
|
87 |
dbc.Col([
|
@@ -166,83 +255,6 @@ app.layout = dbc.Container([
|
|
166 |
])
|
167 |
], fluid=True, style={'padding': '0'})
|
168 |
|
169 |
-
|
170 |
-
# --- File Parsing ---
|
171 |
-
def parse_file_content(file_path, filename):
|
172 |
-
try:
|
173 |
-
with open(file_path, "rb") as f:
|
174 |
-
decoded = f.read()
|
175 |
-
if filename.endswith('.pdf'):
|
176 |
-
with io.BytesIO(decoded) as pdf_file:
|
177 |
-
reader = PyPDF2.PdfReader(pdf_file)
|
178 |
-
return ' '.join([page.extract_text() or "" for page in reader.pages])
|
179 |
-
elif filename.endswith('.docx'):
|
180 |
-
with io.BytesIO(decoded) as docx_file:
|
181 |
-
doc = docx.Document(docx_file)
|
182 |
-
return ' '.join([para.text for para in doc.paragraphs])
|
183 |
-
elif filename.endswith('.txt') or filename.endswith('.rtf'):
|
184 |
-
encoding = chardet.detect(decoded)['encoding']
|
185 |
-
return decoded.decode(encoding)
|
186 |
-
else:
|
187 |
-
return "Unsupported file format"
|
188 |
-
except Exception as e:
|
189 |
-
logger.exception(f"Error processing file {filename}: {str(e)}")
|
190 |
-
return "Error processing file"
|
191 |
-
|
192 |
-
def truncate_filename(filename, max_length=24):
|
193 |
-
if len(filename) <= max_length:
|
194 |
-
return filename
|
195 |
-
else:
|
196 |
-
return filename[:max_length - 3] + '...'
|
197 |
-
|
198 |
-
def get_file_cards(file_dict):
|
199 |
-
cards = []
|
200 |
-
for name in file_dict:
|
201 |
-
cards.append(
|
202 |
-
dbc.Card(
|
203 |
-
dbc.CardBody(
|
204 |
-
dbc.Row([
|
205 |
-
dbc.Col(
|
206 |
-
html.Span(
|
207 |
-
truncate_filename(name),
|
208 |
-
title=name,
|
209 |
-
style={
|
210 |
-
'display': 'inline-block',
|
211 |
-
'overflow': 'hidden',
|
212 |
-
'textOverflow': 'ellipsis',
|
213 |
-
'whiteSpace': 'nowrap',
|
214 |
-
'maxWidth': '90%',
|
215 |
-
'verticalAlign': 'middle',
|
216 |
-
}
|
217 |
-
),
|
218 |
-
width='auto',
|
219 |
-
style={'display': 'flex', 'alignItems': 'center', 'padding': '0'}
|
220 |
-
),
|
221 |
-
dbc.Col(
|
222 |
-
dbc.Button(
|
223 |
-
"Delete",
|
224 |
-
id={'type': 'remove-file', 'index': name},
|
225 |
-
color="danger",
|
226 |
-
size="sm",
|
227 |
-
style={'marginLeft': 'auto', 'float': 'right'}
|
228 |
-
),
|
229 |
-
width='auto',
|
230 |
-
style={'display': 'flex', 'alignItems': 'center', 'justifyContent': 'flex-end', 'padding': '0'}
|
231 |
-
),
|
232 |
-
],
|
233 |
-
justify="between",
|
234 |
-
align="center",
|
235 |
-
style={"margin": "0", "padding": "0"}
|
236 |
-
),
|
237 |
-
style={'padding': '6px 8px', 'margin': '0', 'display': 'flex', 'alignItems': 'center', 'background': 'none', 'boxShadow': 'none'}
|
238 |
-
),
|
239 |
-
style={'border': 'none', 'boxShadow': 'none', 'background': 'none', 'marginBottom': '2px'}
|
240 |
-
)
|
241 |
-
)
|
242 |
-
return cards
|
243 |
-
|
244 |
-
# --- Callbacks ---
|
245 |
-
|
246 |
@app.callback(
|
247 |
Output('file-list', 'children'),
|
248 |
Input('upload-files', 'contents'),
|
@@ -269,12 +281,11 @@ def update_output(list_of_contents, list_of_names):
|
|
269 |
@app.callback(
|
270 |
Output('file-list', 'children', allow_duplicate=True),
|
271 |
Input({'type': 'remove-file', 'index': ALL}, 'n_clicks'),
|
272 |
-
State('file-list', 'children'),
|
273 |
prevent_initial_call=True
|
274 |
)
|
275 |
-
def remove_file(n_clicks
|
276 |
-
session_data, lock = get_session_data()
|
277 |
ctx = dash.callback_context
|
|
|
278 |
if not ctx.triggered:
|
279 |
raise PreventUpdate
|
280 |
triggered_id = ctx.triggered[0]['prop_id'].split('.')[0]
|
@@ -352,7 +363,7 @@ Now, generate the {matrix_type}:
|
|
352 |
]
|
353 |
)
|
354 |
matrix_text = response.choices[0].message.content.strip()
|
355 |
-
logger.info(f"Raw matrix text from GPT: {matrix_text[:200]}...")
|
356 |
lines = [line.strip() for line in matrix_text.split('\n') if '|' in line]
|
357 |
data = [line.split('|') for line in lines]
|
358 |
data = [[cell.strip() for cell in row] for row in data]
|
|
|
12 |
import tempfile
|
13 |
import shutil
|
14 |
import logging
|
15 |
+
from flask import request, make_response, g
|
16 |
from dash.exceptions import PreventUpdate
|
17 |
import PyPDF2
|
18 |
import docx
|
19 |
import chardet
|
20 |
|
|
|
21 |
logging.basicConfig(level=logging.INFO)
|
22 |
logger = logging.getLogger("maiko_matrix_app")
|
23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
SESSION_DATA = {}
|
25 |
SESSION_LOCKS = {}
|
26 |
|
27 |
+
def get_session_id(set_cookie=False):
|
28 |
+
# Try to get session id from browser cookies
|
29 |
+
sid = request.cookies.get('session-id')
|
30 |
+
if sid and sid in SESSION_DATA:
|
31 |
+
return sid
|
32 |
+
# If not present, generate a new one
|
33 |
+
sid = str(uuid.uuid4())
|
34 |
+
if set_cookie:
|
35 |
+
g.set_cookie_sid = sid
|
36 |
+
return sid
|
37 |
|
38 |
def get_session_data():
|
39 |
+
sid = get_session_id()
|
40 |
+
if sid not in SESSION_DATA:
|
41 |
+
SESSION_DATA[sid] = {
|
42 |
+
'uploaded_files': {},
|
43 |
+
'file_texts': {},
|
44 |
'current_matrix': None,
|
45 |
'matrix_type': None,
|
46 |
+
'temp_dir': tempfile.mkdtemp(prefix=f"maiko_{sid}_"),
|
47 |
}
|
48 |
+
SESSION_LOCKS[sid] = threading.Lock()
|
49 |
+
return SESSION_DATA[sid], SESSION_LOCKS[sid]
|
50 |
|
51 |
def cleanup_session_tempdirs():
|
|
|
52 |
for sess in SESSION_DATA.values():
|
53 |
try:
|
54 |
shutil.rmtree(sess['temp_dir'])
|
|
|
75 |
"SWOT Matrix": "Create a matrix analyzing Strengths, Weaknesses, Opportunities, and Threats."
|
76 |
}
|
77 |
|
78 |
+
app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])
|
79 |
+
server = app.server
|
80 |
+
|
81 |
+
openai.api_key = os.environ.get('OPENAI_API_KEY')
|
82 |
+
|
83 |
+
@server.before_request
|
84 |
+
def ensure_session_cookie():
|
85 |
+
# Set session cookie if not present
|
86 |
+
sid = request.cookies.get('session-id')
|
87 |
+
if not sid or sid not in SESSION_DATA:
|
88 |
+
sid_new = str(uuid.uuid4())
|
89 |
+
g.set_cookie_sid = sid_new
|
90 |
+
get_session_id(set_cookie=True)
|
91 |
+
else:
|
92 |
+
g.set_cookie_sid = None
|
93 |
+
|
94 |
+
@server.after_request
|
95 |
+
def set_session_cookie(response):
|
96 |
+
sid = getattr(g, 'set_cookie_sid', None)
|
97 |
+
if sid:
|
98 |
+
response.set_cookie('session-id', sid, max_age=60*60*48, httponly=True)
|
99 |
+
return response
|
100 |
+
|
101 |
+
def parse_file_content(file_path, filename):
|
102 |
+
try:
|
103 |
+
with open(file_path, "rb") as f:
|
104 |
+
decoded = f.read()
|
105 |
+
if filename.endswith('.pdf'):
|
106 |
+
with io.BytesIO(decoded) as pdf_file:
|
107 |
+
reader = PyPDF2.PdfReader(pdf_file)
|
108 |
+
return ' '.join([page.extract_text() or "" for page in reader.pages])
|
109 |
+
elif filename.endswith('.docx'):
|
110 |
+
with io.BytesIO(decoded) as docx_file:
|
111 |
+
doc = docx.Document(docx_file)
|
112 |
+
return ' '.join([para.text for para in doc.paragraphs])
|
113 |
+
elif filename.endswith('.txt') or filename.endswith('.rtf'):
|
114 |
+
encoding = chardet.detect(decoded)['encoding']
|
115 |
+
return decoded.decode(encoding)
|
116 |
+
else:
|
117 |
+
return "Unsupported file format"
|
118 |
+
except Exception as e:
|
119 |
+
logger.exception(f"Error processing file {filename}: {str(e)}")
|
120 |
+
return "Error processing file"
|
121 |
+
|
122 |
+
def truncate_filename(filename, max_length=24):
|
123 |
+
if len(filename) <= max_length:
|
124 |
+
return filename
|
125 |
+
else:
|
126 |
+
return filename[:max_length - 3] + '...'
|
127 |
+
|
128 |
+
def get_file_cards(file_dict):
|
129 |
+
cards = []
|
130 |
+
for name in file_dict:
|
131 |
+
cards.append(
|
132 |
+
dbc.Card(
|
133 |
+
dbc.CardBody(
|
134 |
+
dbc.Row([
|
135 |
+
dbc.Col(
|
136 |
+
html.Span(
|
137 |
+
truncate_filename(name),
|
138 |
+
title=name,
|
139 |
+
style={
|
140 |
+
'display': 'inline-block',
|
141 |
+
'overflow': 'hidden',
|
142 |
+
'textOverflow': 'ellipsis',
|
143 |
+
'whiteSpace': 'nowrap',
|
144 |
+
'maxWidth': '90%',
|
145 |
+
'verticalAlign': 'middle',
|
146 |
+
}
|
147 |
+
),
|
148 |
+
width='auto',
|
149 |
+
style={'display': 'flex', 'alignItems': 'center', 'padding': '0'}
|
150 |
+
),
|
151 |
+
dbc.Col(
|
152 |
+
dbc.Button(
|
153 |
+
"Delete",
|
154 |
+
id={'type': 'remove-file', 'index': name},
|
155 |
+
color="danger",
|
156 |
+
size="sm",
|
157 |
+
style={'marginLeft': 'auto', 'float': 'right'}
|
158 |
+
),
|
159 |
+
width='auto',
|
160 |
+
style={'display': 'flex', 'alignItems': 'center', 'justifyContent': 'flex-end', 'padding': '0'}
|
161 |
+
),
|
162 |
+
],
|
163 |
+
justify="between",
|
164 |
+
align="center",
|
165 |
+
style={"margin": "0", "padding": "0"}
|
166 |
+
),
|
167 |
+
style={'padding': '6px 8px', 'margin': '0', 'display': 'flex', 'alignItems': 'center', 'background': 'none', 'boxShadow': 'none'}
|
168 |
+
),
|
169 |
+
style={'border': 'none', 'boxShadow': 'none', 'background': 'none', 'marginBottom': '2px'}
|
170 |
+
)
|
171 |
+
)
|
172 |
+
return cards
|
173 |
+
|
174 |
app.layout = dbc.Container([
|
175 |
dbc.Row([
|
176 |
dbc.Col([
|
|
|
255 |
])
|
256 |
], fluid=True, style={'padding': '0'})
|
257 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
258 |
@app.callback(
|
259 |
Output('file-list', 'children'),
|
260 |
Input('upload-files', 'contents'),
|
|
|
281 |
@app.callback(
|
282 |
Output('file-list', 'children', allow_duplicate=True),
|
283 |
Input({'type': 'remove-file', 'index': ALL}, 'n_clicks'),
|
|
|
284 |
prevent_initial_call=True
|
285 |
)
|
286 |
+
def remove_file(n_clicks):
|
|
|
287 |
ctx = dash.callback_context
|
288 |
+
session_data, lock = get_session_data()
|
289 |
if not ctx.triggered:
|
290 |
raise PreventUpdate
|
291 |
triggered_id = ctx.triggered[0]['prop_id'].split('.')[0]
|
|
|
363 |
]
|
364 |
)
|
365 |
matrix_text = response.choices[0].message.content.strip()
|
366 |
+
logger.info(f"Raw matrix text from GPT: {matrix_text[:200]}...")
|
367 |
lines = [line.strip() for line in matrix_text.split('\n') if '|' in line]
|
368 |
data = [line.split('|') for line in lines]
|
369 |
data = [[cell.strip() for cell in row] for row in data]
|