Update app.py
Browse files
app.py
CHANGED
|
@@ -24,9 +24,8 @@ def remove_duplicates(df: pd.DataFrame) -> pd.DataFrame:
|
|
| 24 |
|
| 25 |
|
| 26 |
def compute_keyword_score(text: str, keywords: List[str]) -> int:
|
| 27 |
-
"""Simple keyword ranking: count of keyword hits (case-insensitive)."""
|
| 28 |
text_l = (text or "").lower()
|
| 29 |
-
return sum(text_l.count(k.lower()) for k in keywords if k
|
| 30 |
|
| 31 |
|
| 32 |
# ======================================================
|
|
@@ -100,7 +99,7 @@ def extract_search_parameters(client, prompt: str) -> Dict[str, str]:
|
|
| 100 |
|
| 101 |
|
| 102 |
# ======================================================
|
| 103 |
-
# Job scraping
|
| 104 |
# ======================================================
|
| 105 |
|
| 106 |
@st.cache_data(ttl=3600)
|
|
@@ -125,11 +124,6 @@ def get_indeed_jobs(
|
|
| 125 |
return pd.DataFrame()
|
| 126 |
|
| 127 |
|
| 128 |
-
def get_other_board_stub(board_name: str) -> pd.DataFrame:
|
| 129 |
-
"""Stub for future boards (toggle-safe)."""
|
| 130 |
-
return pd.DataFrame()
|
| 131 |
-
|
| 132 |
-
|
| 133 |
# ======================================================
|
| 134 |
# Streamlit App
|
| 135 |
# ======================================================
|
|
@@ -138,102 +132,142 @@ def main():
|
|
| 138 |
st.set_page_config(page_title="Private Job Search", layout="centered")
|
| 139 |
st.title("📄 Private Job Search, Rank & Download")
|
| 140 |
|
| 141 |
-
# ---
|
|
|
|
|
|
|
| 142 |
job_prompt = st.text_area(
|
| 143 |
"Describe the job you are looking for",
|
| 144 |
-
placeholder="e.g. Civil Engineer, Water Resources, Transportation
|
| 145 |
height=120
|
| 146 |
)
|
| 147 |
|
| 148 |
api_key = st.text_input("Groq API Key", type="password")
|
| 149 |
|
| 150 |
-
# ---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 151 |
st.subheader("Job Boards")
|
| 152 |
-
|
| 153 |
-
|
| 154 |
-
|
| 155 |
-
|
| 156 |
-
|
| 157 |
-
with colb3:
|
| 158 |
-
use_linkedin = st.checkbox("LinkedIn (coming soon)", value=False, disabled=True)
|
| 159 |
-
|
| 160 |
-
# --- Filters ---
|
| 161 |
st.subheader("Filters")
|
|
|
|
| 162 |
posted_within_days = st.slider(
|
| 163 |
"Posted within last (days)",
|
| 164 |
-
min_value=1,
|
|
|
|
|
|
|
| 165 |
)
|
| 166 |
|
| 167 |
radius_km = st.slider(
|
| 168 |
"Search radius (km)",
|
| 169 |
-
min_value=5,
|
|
|
|
|
|
|
|
|
|
| 170 |
)
|
| 171 |
|
| 172 |
-
# ---
|
|
|
|
|
|
|
| 173 |
keywords_raw = st.text_input(
|
| 174 |
"Keyword ranking (comma-separated)",
|
| 175 |
-
placeholder="water, wastewater, stormwater, EPANET
|
| 176 |
)
|
| 177 |
keywords = [k.strip() for k in keywords_raw.split(",") if k.strip()]
|
| 178 |
|
| 179 |
-
# ---
|
|
|
|
|
|
|
| 180 |
send_email = st.checkbox("📧 Send results by email (optional)")
|
| 181 |
email_address = st.text_input("Email address") if send_email else None
|
| 182 |
|
| 183 |
-
# ---
|
| 184 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 185 |
client = groq.Client(api_key=api_key)
|
| 186 |
|
| 187 |
with st.spinner("Understanding your request..."):
|
| 188 |
params = extract_search_parameters(client, job_prompt)
|
| 189 |
|
| 190 |
-
|
| 191 |
-
|
| 192 |
-
|
| 193 |
-
|
| 194 |
-
|
| 195 |
-
|
| 196 |
-
|
| 197 |
-
|
| 198 |
-
|
| 199 |
-
|
| 200 |
-
if not indeed_df.empty:
|
| 201 |
-
indeed_df["source"] = "Indeed"
|
| 202 |
-
all_jobs.append(indeed_df)
|
| 203 |
-
|
| 204 |
-
# Future boards (toggle-safe)
|
| 205 |
-
if use_glassdoor:
|
| 206 |
-
all_jobs.append(get_other_board_stub("Glassdoor"))
|
| 207 |
-
if use_linkedin:
|
| 208 |
-
all_jobs.append(get_other_board_stub("LinkedIn"))
|
| 209 |
-
|
| 210 |
-
if not all_jobs:
|
| 211 |
-
st.warning("No jobs found.")
|
| 212 |
return
|
| 213 |
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
|
| 217 |
-
|
| 218 |
-
|
| 219 |
-
|
| 220 |
-
jobs_df["keyword_score"] = jobs_df.apply(
|
| 221 |
-
lambda r: compute_keyword_score(
|
| 222 |
-
f"{r.get('title','')} {r.get('description','')}",
|
| 223 |
-
keywords
|
| 224 |
-
),
|
| 225 |
-
axis=1
|
| 226 |
)
|
| 227 |
-
else:
|
| 228 |
-
jobs_df["keyword_score"] = 0
|
| 229 |
|
| 230 |
-
|
| 231 |
-
|
| 232 |
-
|
| 233 |
|
| 234 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
|
| 236 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 237 |
csv_data = jobs_df.to_csv(index=False).encode("utf-8")
|
| 238 |
st.download_button(
|
| 239 |
label="⬇️ Download Jobs (CSV)",
|
|
@@ -242,7 +276,9 @@ def main():
|
|
| 242 |
mime="text/csv"
|
| 243 |
)
|
| 244 |
|
| 245 |
-
# ---
|
|
|
|
|
|
|
| 246 |
if send_email:
|
| 247 |
if not email_address:
|
| 248 |
st.warning("Please enter an email address.")
|
|
@@ -256,15 +292,20 @@ def main():
|
|
| 256 |
except Exception as e:
|
| 257 |
st.error(f"Failed to send email: {e}")
|
| 258 |
|
| 259 |
-
# ---
|
| 260 |
-
|
|
|
|
|
|
|
| 261 |
preview_cols = [
|
| 262 |
c for c in [
|
| 263 |
-
"
|
| 264 |
"keyword_score", "date_posted", "job_url"
|
| 265 |
] if c in jobs_df.columns
|
| 266 |
]
|
| 267 |
-
st.dataframe(
|
|
|
|
|
|
|
|
|
|
| 268 |
|
| 269 |
|
| 270 |
if __name__ == "__main__":
|
|
|
|
| 24 |
|
| 25 |
|
| 26 |
def compute_keyword_score(text: str, keywords: List[str]) -> int:
|
|
|
|
| 27 |
text_l = (text or "").lower()
|
| 28 |
+
return sum(text_l.count(k.lower()) for k in keywords if k)
|
| 29 |
|
| 30 |
|
| 31 |
# ======================================================
|
|
|
|
| 99 |
|
| 100 |
|
| 101 |
# ======================================================
|
| 102 |
+
# Job scraping
|
| 103 |
# ======================================================
|
| 104 |
|
| 105 |
@st.cache_data(ttl=3600)
|
|
|
|
| 124 |
return pd.DataFrame()
|
| 125 |
|
| 126 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
# ======================================================
|
| 128 |
# Streamlit App
|
| 129 |
# ======================================================
|
|
|
|
| 132 |
st.set_page_config(page_title="Private Job Search", layout="centered")
|
| 133 |
st.title("📄 Private Job Search, Rank & Download")
|
| 134 |
|
| 135 |
+
# --------------------------------------------------
|
| 136 |
+
# Job description
|
| 137 |
+
# --------------------------------------------------
|
| 138 |
job_prompt = st.text_area(
|
| 139 |
"Describe the job you are looking for",
|
| 140 |
+
placeholder="e.g. Civil Engineer, Water Resources, Transportation",
|
| 141 |
height=120
|
| 142 |
)
|
| 143 |
|
| 144 |
api_key = st.text_input("Groq API Key", type="password")
|
| 145 |
|
| 146 |
+
# --------------------------------------------------
|
| 147 |
+
# City selection
|
| 148 |
+
# --------------------------------------------------
|
| 149 |
+
st.subheader("Location")
|
| 150 |
+
|
| 151 |
+
predefined_cities = [
|
| 152 |
+
"Use AI / Prompt Location",
|
| 153 |
+
"Calgary, AB",
|
| 154 |
+
"Edmonton, AB",
|
| 155 |
+
"Toronto, ON",
|
| 156 |
+
"Vancouver, BC",
|
| 157 |
+
"Mississauga, ON",
|
| 158 |
+
"Brampton, ON",
|
| 159 |
+
"Ottawa, ON",
|
| 160 |
+
"Hamilton, ON",
|
| 161 |
+
"Custom city..."
|
| 162 |
+
]
|
| 163 |
+
|
| 164 |
+
selected_city = st.selectbox("Select city", predefined_cities)
|
| 165 |
+
|
| 166 |
+
custom_city = ""
|
| 167 |
+
if selected_city == "Custom city...":
|
| 168 |
+
custom_city = st.text_input(
|
| 169 |
+
"Enter city (e.g., Red Deer, AB or Surrey, BC)"
|
| 170 |
+
)
|
| 171 |
+
|
| 172 |
+
# --------------------------------------------------
|
| 173 |
+
# Job boards
|
| 174 |
+
# --------------------------------------------------
|
| 175 |
st.subheader("Job Boards")
|
| 176 |
+
use_indeed = st.checkbox("Indeed", value=True)
|
| 177 |
+
|
| 178 |
+
# --------------------------------------------------
|
| 179 |
+
# Filters
|
| 180 |
+
# --------------------------------------------------
|
|
|
|
|
|
|
|
|
|
|
|
|
| 181 |
st.subheader("Filters")
|
| 182 |
+
|
| 183 |
posted_within_days = st.slider(
|
| 184 |
"Posted within last (days)",
|
| 185 |
+
min_value=1,
|
| 186 |
+
max_value=30,
|
| 187 |
+
value=7
|
| 188 |
)
|
| 189 |
|
| 190 |
radius_km = st.slider(
|
| 191 |
"Search radius (km)",
|
| 192 |
+
min_value=5,
|
| 193 |
+
max_value=100,
|
| 194 |
+
value=25,
|
| 195 |
+
step=5
|
| 196 |
)
|
| 197 |
|
| 198 |
+
# --------------------------------------------------
|
| 199 |
+
# Keyword ranking
|
| 200 |
+
# --------------------------------------------------
|
| 201 |
keywords_raw = st.text_input(
|
| 202 |
"Keyword ranking (comma-separated)",
|
| 203 |
+
placeholder="water, wastewater, stormwater, EPANET"
|
| 204 |
)
|
| 205 |
keywords = [k.strip() for k in keywords_raw.split(",") if k.strip()]
|
| 206 |
|
| 207 |
+
# --------------------------------------------------
|
| 208 |
+
# Optional email
|
| 209 |
+
# --------------------------------------------------
|
| 210 |
send_email = st.checkbox("📧 Send results by email (optional)")
|
| 211 |
email_address = st.text_input("Email address") if send_email else None
|
| 212 |
|
| 213 |
+
# --------------------------------------------------
|
| 214 |
+
# Action
|
| 215 |
+
# --------------------------------------------------
|
| 216 |
+
if st.button(
|
| 217 |
+
"🔍 Search Jobs",
|
| 218 |
+
disabled=not job_prompt or not api_key
|
| 219 |
+
):
|
| 220 |
client = groq.Client(api_key=api_key)
|
| 221 |
|
| 222 |
with st.spinner("Understanding your request..."):
|
| 223 |
params = extract_search_parameters(client, job_prompt)
|
| 224 |
|
| 225 |
+
# Resolve final location
|
| 226 |
+
if selected_city == "Use AI / Prompt Location":
|
| 227 |
+
location = params.get("location", "Canada")
|
| 228 |
+
elif selected_city == "Custom city...":
|
| 229 |
+
location = custom_city if custom_city else params.get("location", "Canada")
|
| 230 |
+
else:
|
| 231 |
+
location = selected_city
|
| 232 |
+
|
| 233 |
+
if not use_indeed:
|
| 234 |
+
st.warning("No job boards selected.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
return
|
| 236 |
|
| 237 |
+
with st.spinner("Searching jobs..."):
|
| 238 |
+
jobs_df = get_indeed_jobs(
|
| 239 |
+
params["search_term"],
|
| 240 |
+
location,
|
| 241 |
+
radius_km,
|
| 242 |
+
posted_within_days
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 243 |
)
|
|
|
|
|
|
|
| 244 |
|
| 245 |
+
if jobs_df.empty:
|
| 246 |
+
st.warning("No jobs found.")
|
| 247 |
+
return
|
| 248 |
|
| 249 |
+
jobs_df.fillna("", inplace=True)
|
| 250 |
+
jobs_df = remove_duplicates(jobs_df)
|
| 251 |
+
|
| 252 |
+
# Keyword ranking
|
| 253 |
+
jobs_df["keyword_score"] = jobs_df.apply(
|
| 254 |
+
lambda r: compute_keyword_score(
|
| 255 |
+
f"{r.get('title','')} {r.get('description','')}",
|
| 256 |
+
keywords
|
| 257 |
+
),
|
| 258 |
+
axis=1
|
| 259 |
+
)
|
| 260 |
|
| 261 |
+
jobs_df = jobs_df.sort_values(
|
| 262 |
+
by="keyword_score",
|
| 263 |
+
ascending=False
|
| 264 |
+
)
|
| 265 |
+
|
| 266 |
+
st.success(f"✅ Found {len(jobs_df)} jobs for **{location}**")
|
| 267 |
+
|
| 268 |
+
# --------------------------------------------------
|
| 269 |
+
# Download
|
| 270 |
+
# --------------------------------------------------
|
| 271 |
csv_data = jobs_df.to_csv(index=False).encode("utf-8")
|
| 272 |
st.download_button(
|
| 273 |
label="⬇️ Download Jobs (CSV)",
|
|
|
|
| 276 |
mime="text/csv"
|
| 277 |
)
|
| 278 |
|
| 279 |
+
# --------------------------------------------------
|
| 280 |
+
# Optional email
|
| 281 |
+
# --------------------------------------------------
|
| 282 |
if send_email:
|
| 283 |
if not email_address:
|
| 284 |
st.warning("Please enter an email address.")
|
|
|
|
| 292 |
except Exception as e:
|
| 293 |
st.error(f"Failed to send email: {e}")
|
| 294 |
|
| 295 |
+
# --------------------------------------------------
|
| 296 |
+
# Preview
|
| 297 |
+
# --------------------------------------------------
|
| 298 |
+
st.subheader("Preview (Top 20 Results)")
|
| 299 |
preview_cols = [
|
| 300 |
c for c in [
|
| 301 |
+
"title", "company", "location",
|
| 302 |
"keyword_score", "date_posted", "job_url"
|
| 303 |
] if c in jobs_df.columns
|
| 304 |
]
|
| 305 |
+
st.dataframe(
|
| 306 |
+
jobs_df[preview_cols].head(20),
|
| 307 |
+
use_container_width=True
|
| 308 |
+
)
|
| 309 |
|
| 310 |
|
| 311 |
if __name__ == "__main__":
|