Release/v0.1.3

.gitignore

@@ -175,3 +175,6 @@ Rscripts/
 
 # Development
 dev
+
+# Deployment
+deployments.txt

Dockerfile

@@ -33,4 +33,5 @@ RUN uv pip install -e .
 EXPOSE 80
 
 # Start the server
-CMD ["uv", "run", "streamlit", "run", "hydroshift/streamlit_app.py", "--server.port=80", "--server.address=0.0.0.0", "--server.headless=true"]
+CMD uv run hydroshift/add_analytics.py && uv run streamlit run hydroshift/streamlit_app.py \
+    --server.port=80 --server.address=0.0.0.0 --server.headless=true

docker-compose.yml

@@ -8,7 +8,14 @@ services:
     restart: unless-stopped
     mem_limit: 4GB
     mem_reservation: 2GB
-
+    stdin_open: true
+    labels:
+      - 'traefik.enable=true'
+      - 'traefik.http.routers.client.rule=Host(`${PROD_APP_HOST}`)'
+      - 'traefik.http.routers.client.tls=true'
+      - 'traefik.http.routers.client.tls.certresolver=letsencrypt'
+      - 'traefik.docker.network=hydroshift_hydroshift'
+      - 'traefik.http.routers.static.middlewares=secHeaders@file'
 networks:
   hydroshift:
     driver: bridge

hydroshift/__init__.py

@@ -1 +1 @@
-__version__ = "0.1.2"
+__version__ = "0.1.3"

hydroshift/_pages/__init__.py

@@ -1,3 +1,4 @@
-from hydroshift._pages.changepoint import changepoint
-from hydroshift._pages.homepage import homepage, reset_homepage
-from hydroshift._pages.summary import summary
+from hydroshift._pages.changepoint import changepoint as changepoint
+from hydroshift._pages.homepage import homepage as homepage
+from hydroshift._pages.homepage import reset_homepage as reset_homepage
+from hydroshift._pages.summary import summary as summary

hydroshift/_pages/changepoint.py

@@ -72,9 +72,7 @@ def get_change_windows(self, max_dist: float = 10) -> tuple:
         current_group_tests = set(test_dict[dates[0]])
 
         for i in range(1, len(dates)):
-            if (
-                (dates[i] - current_group[-1]).days / 365
-            ) < max_dist:  # 10 years in days
+            if ((dates[i] - current_group[-1]).days / 365) < max_dist:  # 10 years in days
                 current_group.append(dates[i])
                 current_group_tests = current_group_tests.union(test_dict[dates[i]])
             else:
@@ -93,12 +91,7 @@ def get_max_pvalue(self) -> tuple[float, int]:
         """Get the minimum p value where all tests agree and count how often that occurred."""
         all_tests = self.pval_df.fillna(1).max(axis=1).to_frame(name="pval")
         all_tests["run"] = ((all_tests != all_tests.shift(1)) * 1).cumsum()
-        for ind, r in (
-            all_tests.groupby("pval")
-            .agg({"run": pd.Series.nunique})
-            .sort_index()
-            .iterrows()
-        ):
+        for ind, r in all_tests.groupby("pval").agg({"run": pd.Series.nunique}).sort_index().iterrows():
             if r.run > 0:
                 min_p = ind
                 count = r.run
@@ -127,9 +120,7 @@ def ffa_png(self):
     @property
     def cp_df(self):
         """Get a dataframe representing changepoints identified in the streaming analysis."""
-        cpa_df = pd.DataFrame.from_dict(
-            self.cp_dict, orient="index", columns=["Tests Identifying Change"]
-        )
+        cpa_df = pd.DataFrame.from_dict(self.cp_dict, orient="index", columns=["Tests Identifying Change"])
         cpa_df.index = cpa_df.index.date
         return cpa_df
 
@@ -184,9 +175,7 @@ def results_text(self) -> str:
             "plural": p_count > 1,
             "len_cp": len(self.cp_dict),
             "len_cp_str": num_2_word(len(self.cp_dict)),
-            "test_count": num_2_word(
-                len(self.cp_dict[next(iter(self.cp_dict))].split(","))
-            ),
+            "test_count": num_2_word(len(self.cp_dict[next(iter(self.cp_dict))].split(","))),
             "grp_count": num_2_word(len(groups)),
             "plural_2": len(groups) > 0,
         }
@@ -222,9 +211,7 @@ def word_data(self) -> BytesIO:
             document.add_heading("Modified flood frequency analysis", level=2)
             self.add_markdown_to_doc(document, self.ffa_text)
             document.add_picture(self.ffa_png, width=Inches(6.5))
-            self.add_markdown_to_doc(
-                document, "**Figure 2.** Modified flood frequency analysis."
-            )
+            self.add_markdown_to_doc(document, "**Figure 2.** Modified flood frequency analysis.")
             self.add_markdown_to_doc(document, "**Table 2.** Modified flood quantiles.")
             self.add_table_from_df(self.ffa_df, document, index_name="Regime Period")
 
@@ -235,9 +222,7 @@ def word_data(self) -> BytesIO:
         document.save(out)
         return out
 
-    def add_table_from_df(
-        self, df: pd.DataFrame, document: Document, index_name: str = None
-    ):
+    def add_table_from_df(self, df: pd.DataFrame, document: Document, index_name: str = None):
         if index_name is not None:
             df = df.copy()
             cols = df.columns
@@ -269,10 +254,12 @@ def validate_data(self):
 
         # Validate
         if data is None:
-            return False,  "Unable to retrieve data."
+            return False, "Unable to retrieve data."
         elif len(data) < st.session_state.burn_in:
             st.session_state.valid_data = False
-            return False, "Not enough peaks available for analysis. {} peaks found, but burn-in length was {}".format(len(data["peak_va"]), st.session_state.burn_in)
+            return False, "Not enough peaks available for analysis. {} peaks found, but burn-in length was {}".format(
+                len(data["peak_va"]), st.session_state.burn_in
+            )
         else:
             return True, None
 
@@ -336,9 +323,7 @@ def make_sidebar():
             # Make data editor.  Unique key allows for refreshing
             if "data_editor_key" not in st.session_state:
                 refresh_data_editor()
-            start_config = st.column_config.DateColumn(
-                "Regime Start", format="D/M/YYYY"
-            )
+            start_config = st.column_config.DateColumn("Regime Start", format="D/M/YYYY")
             end_config = st.column_config.DateColumn("Regime End", format="D/M/YYYY")
             st.data_editor(
                 init_data,
@@ -351,13 +336,12 @@ def make_sidebar():
         st.divider()
         write_template("data_sources_side_bar.html")
 
+
 def run_analysis():
     """Run the change point model analysis."""
     cpa = st.session_state.changepoint
     cpa.pval_df = get_pvalues(cpa.gage.ams)
-    cpa.cp_dict = get_changepoints(
-        cpa.gage.ams, st.session_state.arlo_slider, st.session_state.burn_in
-    )
+    cpa.cp_dict = get_changepoints(cpa.gage.ams, st.session_state.arlo_slider, st.session_state.burn_in)
 
 
 @st.cache_data(max_entries=MAX_CACHE_ENTRIES)
@@ -392,7 +376,7 @@ def ffa_analysis(data: pd.DataFrame, regimes: list):
         if "Regime Start" in r and "Regime End" in r:
             sub = data.loc[r["Regime Start"] : r["Regime End"]].copy()
             peaks = sub["peak_va"].values
-            label = f'{r["Regime Start"]} - {r["Regime End"]}'
+            label = f"{r['Regime Start']} - {r['Regime End']}"
             lp3 = LP3Analysis(
                 st.session_state.gage_id,
                 peaks,
@@ -464,10 +448,7 @@ def make_body():
 
 def warnings():
     """Print warnings on data validity etc."""
-    if (
-        st.session_state.changepoint.gage.ams is not None
-        and "peak_va" in st.session_state.changepoint.gage.ams.columns
-    ):
+    if st.session_state.changepoint.gage.ams is not None and "peak_va" in st.session_state.changepoint.gage.ams.columns:
         if st.session_state.changepoint.gage.missing_dates_ams:
             st.warning(
                 "Missing {} dates between {} and {}".format(

hydroshift/_pages/homepage.py

@@ -1,24 +1,11 @@
 import streamlit as st
+
 from hydroshift._pages import summary
 from hydroshift.consts import DEFAULT_GAGE
 from hydroshift.utils.jinja import write_template
 
-
-def homepage():
-    """Landing page for app."""
-    # st.session_state["gage_id"] = None
-    st.set_page_config(layout="centered", initial_sidebar_state ="collapsed")
-
-    st.markdown(
-        """
+PAGE_CSS = """
         <style>
-        .stApp {
-        background: linear-gradient(
-            #f5f5f5 0%,
-            #f5f5f5 45%,
-            #b3c7e8 100%
-        );
-        }
         .stAppDeployButton {display:none;}
         .stAppHeader {display:none;}
         .block-container {
@@ -38,24 +25,59 @@ def homepage():
             box-shadow: 0px 4px 12px rgba(0,0,0,0.25);
             color: black;
         }
+        .stApp {
+            background: linear-gradient(
+                #f5f5f5 0%,
+                #f5f5f5 45%,
+                #b3c7e8 100%
+            ) !important;
+        }
+
+        @media (prefers-color-scheme: dark) {
+            .stApp {
+                background: linear-gradient(
+                    #05051c 0%,
+                    #05051c 45%,
+                    #17428a 100%
+                ) !important;
+            }
+            a {
+                color: #d4d4d4 !important;
+                text-decoration: none;
+            }
+            a:hover {
+                color: #808080 !important;
+                text-decoration: underline;
+            }
+            p {
+                color: #d4d4d4 !important;
+            }
+        }
         </style>
-        """,
+        """
+
+
+def homepage():
+    """Landing page for app."""
+    st.set_page_config(layout="centered", initial_sidebar_state="collapsed")
+
+    st.markdown(
+        PAGE_CSS,
         unsafe_allow_html=True,
     )
 
-    # --- Centered content ---
-    with st.container(horizontal_alignment ="center"):
+    with st.container(horizontal_alignment="center"):
         st.title("HydroShift")
 
-        with st.container(horizontal=True, horizontal_alignment ="center"):
+        with st.container(horizontal=True, horizontal_alignment="center"):
             st.image("hydroshift/images/logo_base.png", width=400)
 
         st.subheader("USGS Streamflow Change Detection Tool")
         write_template("app_description.md")
 
         gage_input = st.text_input("Enter a USGS Gage Number:", placeholder="e.g., 01646500")
 
-        with st.container(horizontal=True, horizontal_alignment ="center"):
+        with st.container(horizontal=True, horizontal_alignment="center"):
             submit = st.button("Submit")
             demo = st.button("Use Demo Data")
 
@@ -70,6 +92,7 @@ def homepage():
 
     write_template("footer.html")
 
+
 def reset_homepage():
     st.session_state["gage_id"] = None
     st.rerun()

hydroshift/_pages/summary.py

@@ -1,6 +1,5 @@
-from datetime import date
 import datetime
-import time
+from datetime import date
 
 import folium
 import streamlit as st
@@ -137,14 +136,13 @@ def section_monthly_mean(gage: Gage):
     "Log-Pearson III (LP3) Analysis": section_lp3,
     "AMS Seasonal Ranking": section_ams_seasonal,
     "Daily Mean Streamflow": section_daily_mean,
-    "Monthly Mean Streamflow": section_monthly_mean
+    "Monthly Mean Streamflow": section_monthly_mean,
 }
 
 
-
 def summary():
     """Display summary plots for various timeseries associated with this gage."""
-    st.set_page_config(page_title="Gage Summary", layout="wide", initial_sidebar_state ="auto")
+    st.set_page_config(page_title="Gage Summary", layout="wide", initial_sidebar_state="auto")
 
     # Sidebar for input
     with st.sidebar:
@@ -180,7 +178,10 @@ def summary():
                     # Display site metadata
                     st.subheader("Site Information")
                     write_template("site_summary.md", gage.site_data)
-                    st.link_button("Go to USGS", f'https://waterdata.usgs.gov/monitoring-location/USGS-{st.session_state["gage_id"]}/')
+                    st.link_button(
+                        "Go to USGS",
+                        f"https://waterdata.usgs.gov/monitoring-location/USGS-{st.session_state['gage_id']}/",
+                    )
 
         with st.spinner():
             gage.raise_warnings()

hydroshift/add_analytics.py

@@ -0,0 +1,34 @@
+import os
+from bs4 import BeautifulSoup
+import pathlib
+import shutil
+import streamlit as st
+
+GA_ID = "google_analytics"
+GA_TAG = os.getenv("GA_TAG", "")  # pull tag from env, fallback to empty
+GA_SCRIPT = f"""
+<!-- Google tag (gtag.js) -->
+<script async src="https://www.googletagmanager.com/gtag/js?id={GA_TAG}"></script>
+<script id="{GA_ID}">
+  window.dataLayer = window.dataLayer || [];
+  function gtag(){{dataLayer.push(arguments);}}
+  gtag('js', new Date());
+
+  gtag('config', '{GA_TAG}');
+</script>
+"""
+
+def inject_ga():
+    index_path = pathlib.Path(st.__file__).parent / "static" / "index.html"
+    soup = BeautifulSoup(index_path.read_text(), features="html.parser")
+    if not soup.find(id=GA_ID):
+        bck_index = index_path.with_suffix('.bck')
+        if bck_index.exists():
+            shutil.copy(bck_index, index_path)
+        else:
+            shutil.copy(index_path, bck_index)
+        html = str(soup)
+        new_html = html.replace('<head>', '<head>\n' + GA_SCRIPT)
+        index_path.write_text(new_html)
+
+inject_ga()

hydroshift/logging.py → hydroshift/app_logging.py

@@ -1,8 +1,8 @@
 # logging_config.py
 import logging
 import logging.handlers
-from pathlib import Path
 import sys
+from pathlib import Path
 
 
 def handle_uncaught(exc_type, exc_value, exc_traceback):
@@ -18,7 +18,7 @@ def setup_logging(log_dir: str = "logs", log_level: int = logging.INFO, log_file
     log_path = Path(log_dir) / log_file
 
     # Establish format
-    log_format = ("%(asctime)s | %(levelname)-8s | %(name)s.%(funcName)s:%(lineno)d | %(message)s")
+    log_format = "%(asctime)s | %(levelname)-8s | %(name)s.%(funcName)s:%(lineno)d | %(message)s"
     formatter = logging.Formatter(log_format, "%Y-%m-%d %H:%M:%S")
 
     # Establish global settings

hydroshift/consts.py

@@ -1,4 +1,5 @@
 """Shared variables."""
+
 # Performance
 MAX_CACHE_ENTRIES = 25
 
@@ -60,13 +61,16 @@
     "C": "All or part of the record affected by Urbanization, Mining, Agricultural changes, Channelization, or other",
 }
 
+
 ### IMAGES ###
 def svg2text(path: str) -> str:
     with open(path, "r") as f:
         svg = f.read()
     if svg.startswith("<?xml"):
         svg = svg.split("?>", 1)[1]
     return svg.strip()
+
+
 GITHUB_SVG = svg2text("hydroshift/images/github_logo.svg")
 DEWBERRY_SVG = svg2text("hydroshift/images/dewberry_logo.svg")
 MAIL_SVG = svg2text("hydroshift/images/mail_logo.svg")