upload

app.py

@@ -10,5 +10,188 @@ import tempfile
 from io import BytesIO
 
 
-x = st.slider('Select a value')
-st.write(x, 'squared is', x * x)
+
+def convert_to_gdf(uploaded_file):
+    # Read the file using BytesIO
+    file_buffer = BytesIO(uploaded_file.read())
+
+    # Detect file type and load accordingly
+    if uploaded_file.name.endswith('.shp'):
+        gdf = gpd.read_file(file_buffer)
+    elif uploaded_file.name.endswith(('.geojson', '.json')):
+        gdf = gpd.read_file(file_buffer, driver='GeoJSON')
+    else:
+        raise ValueError("Unsupported file format")
+
+    return gdf
+
+# add logo D:\Terradot\repos\crea-carbon-model\app\logo.jpg
+
+st.sidebar.image('logo.jpg', width=200)
+st.sidebar.title('Proyecto Crea')
+st.sidebar.write('Solo uso interno')
+# add sidebar with 2 upload buttons
+st.sidebar.header('Upload Files')
+uploaded_file = st.sidebar.file_uploader('Upload your shapefile', type=['shp', 'geojson', 'json'])
+uploaded_file2 = st.sidebar.file_uploader('Upload your csv file', type=['csv'])
+
+if uploaded_file is not None:
+        lotes_gdf = convert_to_gdf(uploaded_file)
+        st.write(lotes_gdf)
+
+if uploaded_file2 is not None:
+    # read csv and create dataframe
+    obs_df_2023 = pd.read_csv(uploaded_file2)
+
+
+# add Test button
+test = True #st.sidebar.button('Test')
+
+
+
+
+if test:
+
+    lotes_gdf = gpd.read_file('data/lotes espacio crea_empresa.shp', encoding='utf-8')
+    obs_df_2023 = pd.read_csv('data/obs_df_2023_12_1.csv')
+    obs_df_2023.fillna('-', inplace=True)
+    obs_df_2023.Campo = obs_df_2023.Campo.astype('str')
+
+    empresa_obs = obs_df_2023.EMPRESA.unique().tolist()
+    # create a state variable to hold the current value of key variable
+
+    if 'key' not in st.session_state:
+        st.session_state['key'] = None
+
+    col1, col2,col3 = st.columns(3)
+
+    with col1:
+        st.header('EMPRESA')
+        selected_company = st.selectbox(f'Seleccione empresa', empresa_obs, index= 0)
+
+    # filter dataframe by selected company
+    obs_df_2023 = obs_df_2023[obs_df_2023['EMPRESA'] == selected_company]
+    lotes_gdf = lotes_gdf[lotes_gdf['empresa'] == selected_company]
+
+    campo_obs = obs_df_2023.Campo.unique().tolist()
+    campo_gdf = lotes_gdf.campo.unique().tolist()
+    # Initialize an empty dictionary
+    similar_dict = {}
+    N = 3
+    CUTOFF = 0.72
+    # Loop through each item in the template list
+    for item in campo_gdf:
+        # normalize the stings to lowercase and remove punctuation in campo_obs
+        campo_obs_norm = [str(c).lower() for c in campo_obs]
+        campo_obs_norm = [c.replace('.', ' ') for c in campo_obs_norm]
+
+        # Find the most similar item in df_columns list
+        similar_items = get_close_matches(item, campo_obs_norm, N, CUTOFF)
+
+        # get the index of the most similar item
+        similar_items_idx = [campo_obs_norm.index(i) for i in similar_items]
+
+        # get the most similar item in the original list
+        similar_items = [campo_obs[i] for i in similar_items_idx]
+
+        # If a similar item is found, add to the dictionary
+        if similar_items:
+            similar_dict[item] = similar_items[0]
+        else:
+            # If no similar item is found, set value as "no match"
+            similar_dict[item] = "no match"
+
+
+
+    similar_dict_df = pd.DataFrame.from_dict(similar_dict, orient='index').reset_index()
+    similar_dict_df.columns = ['gdf','obs']
+    # campo_obs = [str(c) for c in campo_obs]
+
+
+    # campo_obs.sort(key=str.lower)
+    campo_obs.insert(0, 'no match')
+
+    all_keys = similar_dict_df['gdf'].unique().tolist()
+
+    # Fields
+
+    lotes_gdf['campo_obs'] = lotes_gdf['campo'].map(similar_dict)
+
+    cutoff = 0.3
+    for campo , df in lotes_gdf.groupby('campo'):
+        for i, row in df.iterrows():
+            c_obs = obs_df_2023.loc[obs_df_2023.Campo == row.campo_obs]
+            lote_obs = c_obs.Lote.unique()
+            # normalize the stings to lowercase and remove punctuation
+            lote_obs_norm = [str(c).lower() for c in lote_obs]
+            lote_obs_norm = [c.replace('.', ' ') for c in lote_obs_norm]
+            similar_items = get_close_matches(row.lote, lote_obs_norm, N, cutoff)
+            # get the index of the most similar item
+            similar_items_idx = [lote_obs_norm.index(i) for i in similar_items]
+
+            # get the most similar item in the original list
+            similar_items = [lote_obs[i] for i in similar_items_idx]
+
+            if similar_items:
+                lotes_gdf.loc[i, 'lote_obs'] = similar_items[0]
+            else:
+                lotes_gdf.loc[i, 'lote_obs'] = "no match"
+
+
+    def show_field(key):
+        with col3:
+            # st.header(st.session_state['key'])
+            st.header('Lote')
+            df_field = lotes_gdf[lotes_gdf['campo'] == st.session_state['key']]
+            fields = df_field['lote'].unique().tolist()
+            fields_obs = df_field['lote_obs'].unique().tolist()
+            with st.form("fields_form"):
+                for j,field in enumerate(fields):
+                    default = df_field[df_field['lote']==field]['lote_obs'].values[0]
+                    selected_value = st.multiselect(f'{field} (.shp):', fields_obs, default=default, key='field'+str(j))
+
+                if  st.form_submit_button("Submit"):
+                    for j,field in enumerate(fields):
+                        if selected_value:
+                            lotes_gdf.loc[(lotes_gdf['campo'] == st.session_state['key']) & (lotes_gdf['lote'] == field), 'lote_obs'] = selected_value[j]
+                        else:
+                            lotes_gdf.loc[(lotes_gdf['campo'] == st.session_state['key']) & (lotes_gdf['lote'] == field), 'lote_obs'] = 'no match'
+                
+                # st.dataframe(df.drop(columns=['geometry','nombre','apellido','campo','campo_obs']))
+
+
+    def on_click(key):
+        def inner():
+            st.session_state['key'] = key
+            show_field(key)
+        return inner
+
+    with col2:
+        st.header('Campo')
+        for i, key in enumerate(all_keys):
+            # selected_value = st.selectbox(f'{key}:', campo_obs, index=campo_obs.index(similar_dict_df[similar_dict_df['gdf'] == key]['obs'].values[0]))
+            selected_value = st.multiselect(f'{key} (.shp):', campo_obs, default=similar_dict_df[similar_dict_df['gdf'] == key]['obs'].values[0], key=i)
+            if selected_value:
+                similar_dict_df.loc[similar_dict_df['gdf'] == key, 'obs'] = selected_value[0]
+            else:
+                similar_dict_df.loc[similar_dict_df['gdf'] == key, 'obs'] = 'no match'
+
+            st.button('Show Fields', key=key, on_click=on_click(key))
+
+
+        # st.dataframe(similar_dict_df)
+        # if st.button('Show Fields'):
+        #     st.dataframe(similar_dict_df)
+
+    # add download button
+            
+
+            
+
+st.sidebar.download_button(
+        label="Download GeoJSON",
+        data=lotes_gdf.to_json().encode('utf-8'),
+        file_name='lotes.geojson',
+        # mime='text/csv',
+        mime = 'application/json',
+    )

data/lotes espacio crea_empresa.prj

@@ -0,0 +1 @@
+GEOGCS["GCS_WGS_1984",DATUM["D_WGS_1984",SPHEROID["WGS_1984",6378137.0,298.257223563]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]]