| import ast |
| import gradio as gr |
| from pycspwrapper import LVStat |
| import requests |
| from typing import Any, Dict, List, Tuple |
|
|
|
|
| def get_topics(name:str = '') -> dict: |
| """Available topics from Official Statistics Portal of Latvia (CSP or Centrālā statistikas pārvalde). |
| |
| Args: |
| name (str): name of the topic. If not defined, function will return all available topics. |
| Returns: |
| dict: The dictionary of topics, where key is topic name and value is topic code. |
| Examples: |
| >>> get_topics('vide') |
| {'Vide': 'ENV'} |
| >>> print(get_topics()) |
| {'Iedzīvotāji': 'POP', 'Darbs': 'EMP', 'Sociālā aizsardzība un veselība': 'VES',... |
| """ |
| name_capit = name.capitalize() |
| base_url = 'https://data.stat.gov.lv/api/v1/lv/OSP_PUB/' |
| content = requests.get(base_url) |
| content_short = {i['text']: i['id'] for i in content.json()} |
|
|
| if name_capit in content_short.keys(): |
| return {name_capit: content_short[name_capit]} |
| else: |
| return content_short |
|
|
|
|
| def get_topic_content(topic: str) -> dict: |
| """Available contents of the topic from Official Statistics Portal of Latvia (CSP or Centrālā statistikas pārvalde). |
| |
| Args: |
| topic (str): topic code. Use get_topics to get topic code. |
| Returns: |
| dict: The dictionary of the contents of the topic, where key is the topic content and value is the topic content code. |
| Examples: |
| >>> # First get topic code |
| ... get_topics('vide') |
| {'Vide': 'ENV'} |
| >>> # Then use this code to get content |
| ... print(get_topic_content('ENV')) |
| {'Vides konti': 'VI', 'Atkritumu apsaimniekošana': 'AK', 'Agro-vides rādītāji': 'AV',... |
| >>> get_topics('Iedzīvotāji') |
| {'Iedzīvotāji': 'POP'} |
| >>> print(get_topic_content('POP')) |
| {'Iedzīvotāju skaits un raksturojošie rādītāji': 'IR', 'Dzimstība': 'ID', 'Mirstība': 'IM', 'Nāves cēloņi': 'NC',... |
| """ |
| base_url = 'https://data.stat.gov.lv/api/v1/lv/OSP_PUB/START/' |
| content = requests.get(base_url+topic.upper()) |
| content_short = {i['text']: i['id'] for i in content.json()} |
| return content_short |
|
|
|
|
| def get_titles(topic_content_code:str = '', |
| url:str = 'https://data.stat.gov.lv/api/v1/lv/OSP_PUB?query=*&filter=*') -> dict: |
| """Available data (titles) from Official Statistics Portal of Latvia (CSP or Centrālā statistikas pārvalde). |
| |
| Args: |
| topic_content_code (str): topic content code. Use get_topic_content to get topic content code. |
| If not defined, function will return all available titles. |
| url (str): URL from where to get list of available titles. Default value: 'https://data.stat.gov.lv/api/v1/lv/OSP_PUB?query=*&filter=*'. |
| Returns: |
| dict: The dictionary of the titles available from Official Statistics Portal, |
| where key is the title name and value is the list of 4 elements: topic code, topic content code, topic sub-content code and report ID. |
| Examples: |
| >>> # First get topic code |
| ... get_topics('Darbs') |
| {'Darbs': 'EMP'} |
| >>> # Then use this code to get content |
| ... print(get_topic_content('EMP')) |
| {'Darba samaksa (algas)': 'DS', 'Darbaspēka izmaksas': 'DI', 'Darbvietas un darba laiks': 'DV',... |
| >>> # Then use this content code to extract report titles |
| ... print(get_titles('DS')) |
| {'Dzīvi un nedzīvi dzimušo skaits pēc dzimuma 1920 - 2020': ['POP', 'ID', 'IDS', 'IDS010'],... |
| """ |
| alldb = requests.get(url) |
| dict_result = {} |
| for i in alldb.json(): |
| if topic_content_code == i['path'].split('/')[-2]: |
| dict_result[i['title']] = [j for j in i['path'].split('/') if j]+[i['id']] |
| return dict_result |
|
|
| def get_query_values(topic_params: list[str] = []) -> List[Dict]: |
| """Get query code and values for particular report. |
| |
| Args: |
| topic_params (list[str]): arguments as a list that are needed for data extraction. |
| Arguments in the list should be in the following order: |
| - topic code, |
| - topic content code, |
| - topic sub-content code |
| - report ID. |
| These codes you can get from the function get_titles. |
| Returns: |
| Dict: A dictionary where each key is a query parameter code and the value is another dictionary mapping possible values to their descriptive texts. |
| Examples: |
| >>> # First get report topic parameters from get_titles |
| ... print(get_titles('DS')) |
| {'Dzīvi un nedzīvi dzimušo skaits pēc dzimuma 1920 - 2020': ['POP', 'ID', 'IDS', 'IDS010'],... |
| >>> # Then use these values to get possible query values |
| ... print(get_query_values(['POP', 'ID', 'IDS', 'IDS010'])) |
| {'SEX_NEWBORN': {'T': 'Pavisam', 'M': 'Vīrieši', 'F': 'Sievietes'}, 'ContentsCode': {'IDS010': 'Dzīvi dzimuši', 'IDS0101': 'Nedzīvi dzimuši', 'IDS0102': 'Nedzīvi dzimuši uz 1000 dzīvi dzimušiem'}, |
| 'TIME': {'1920': '1920' ... '2024': '2024'}} |
| """ |
| base_url = 'https://data.stat.gov.lv/api/v1/lv/OSP_PUB/START/' |
| url = base_url + '/'.join(topic_params) |
| response = requests.get(url) |
| try: |
| response.raise_for_status() |
| data = response.json() |
| if 'variables' not in data: |
| raise ValueError("Unexpected JSON structure: 'variables' key missing") |
| |
| result = {} |
|
|
| for var in data['variables']: |
| code = var.get('code', '') |
| values = var.get('values', []) |
| value_texts = var.get('valueTexts', []) |
| |
| mapping = dict(zip(values, value_texts)) |
| result[code] = mapping |
| |
| return result |
| except requests.exceptions.RequestException as e: |
| raise RuntimeError(f"Request failed: {e}") |
| except ValueError as ve: |
| raise RuntimeError(f"Parsing failed: {ve}") |
|
|
| def construct_csp_link(params: list[str]) -> str: |
| base_url = 'https://data.stat.gov.lv/pxweb/lv/OSP_PUB/START__' |
| mid_path = '__'.join(params[:3]) |
| last_part = params[3] |
| return f"{base_url}{mid_path}/{last_part}/" |
|
|
| def get_csp_data(lang: str = 'en', topic_params: list[str] = [], **kwargs) -> List[Dict]: |
| """Get statistics from Official Statistics Portal of Latvia (CSP or Centrālā statistikas pārvalde). Use 'Source URL' from the Returns to cite the data source. |
| |
| Args: |
| lang (str): Language. Default value 'en'. |
| topic_params (list[str]): arguments as a list that are needed for data extraction. |
| Arguments in the list should be in the following order: |
| - topic code, |
| - topic content code, |
| - topic sub-content code |
| - report ID. |
| These codes you can get from the function get_titles. |
| kwargs: Keyword arguments for query configuration. |
| Possible query argument names and their possible values |
| can be obtained using the function get_query_values. |
| Returns: |
| list: The list of the dictionaries, where dictionary's key 'key' contains query parameters and key 'values' contains values. First list element is data source (URL) from CSB. |
| Examples: |
| >>> topics = ['POP', 'IR', 'IRE', 'IRE010'] |
| >>> query_args = get_query_values(topics) |
| >>> print(query_args) |
| {'ETHNICITY': {'TOTAL': 'Pavisam', 'E_LAT': 'Latvieši', 'E_ABZ': 'Abāzi', 'E_ABK': 'Abhāzi',... |
| 'E_SWE': 'Zviedri', 'OTH': 'Cita tautība', 'UNK_NSP': 'Nezināma, neizvēlēta'}, 'ContentsCode': {'IRE010': 'Skaits'}, |
| 'TIME': {'1935': '1935', ... '2025': '2025'}} |
| >>> # Then use these codes and values to get data for example Latvians for years 2024 and 2025. |
| >>> # Value text 'Latvieši' explains what 'E_LAT' means. |
| >>> data = get_csp_data( |
| ... lang='en', |
| ... topic_params=topics, |
| ... ETHNICITY=['E_LAT'], |
| ... TIME=['2024', '2025'] |
| ... ) |
| >>> print(data[0]) |
| {'key': ['E_LAT', '2024'], 'values': ['1186337']} |
| """ |
| csp2 = LVStat(lang, *topic_params) |
| csp2.set_query(**kwargs) |
|
|
| link = construct_csp_link(topic_params) |
| |
| cspdata2 = csp2.get_data() |
| return [{"Source URL": link}] + cspdata2['data'] |
|
|
|
|
| with gr.Blocks() as demo: |
| gr.Markdown("### Latvian CSP Data Query Interface") |
| |
| lang = gr.Dropdown(["en", "lv"], value="en", label="Language") |
|
|
| |
| topic_dict = get_topics() |
| topic_dropdown = gr.Dropdown(choices=list(topic_dict.keys()), label="Select Topic") |
|
|
| |
| topic_content_dropdown = gr.Dropdown(label="Select Topic Content", visible=False) |
|
|
| |
| report_dropdown = gr.Dropdown(label="Select Report", visible=False) |
|
|
| |
| link_output = gr.Markdown(visible=False) |
| topic_params_box = gr.Textbox(label="Topic Params", lines=1, interactive=True) |
|
|
| |
| kwargs_box = gr.Textbox( |
| label="Query Parameters (Python dict, e.g., {'ETHNICITY': ['E_LAT'], 'TIME': ['2024', '2025']})", |
| lines=4, |
| ) |
| output = gr.JSON(label="Result") |
|
|
| run_button = gr.Button("Run Query") |
|
|
| def update_topic_content(topic_name: str)-> Tuple[Any, Dict[str, str], str]: |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| topic_code = topic_dict[topic_name] |
| content_dict = get_topic_content(topic_code) |
| return gr.update(choices=list(content_dict.keys()), visible=True), content_dict, topic_code |
|
|
| def update_reports(topic_content_name: str, content_dict: dict = None) -> Tuple[Dict[str, str], str, Any]: |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| topic_content_code = content_dict[topic_content_name] |
| titles_dict = get_titles(topic_content_code) |
| return titles_dict, topic_content_code, gr.update(choices=list(titles_dict.keys()), visible=True) |
|
|
| def update_topic_params_and_link(report_title: str, titles_dict: dict) -> Tuple[str, Any, Any]: |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| title_value = titles_dict[report_title] |
| topic_params_str = str(title_value) |
| link = construct_csp_link(title_value) |
|
|
| try: |
| q = get_query_values(title_value) |
| timeval = q.get('TIME','') |
| query = {} |
|
|
| if len(timeval) > 0: |
| query = {'TIME': list(timeval.keys())[-3:]} |
| |
| |
| |
| |
| |
| query_str = str(query) |
| except Exception as e: |
| query_str = '{}' |
|
|
| return topic_params_str, gr.update(value=f"[{report_title}]({link})", visible=True), gr.update(value=query_str) |
|
|
| topic_content_state = gr.State() |
| titles_state = gr.State() |
| topic_code_state = gr.State() |
| topic_content_code_state = gr.State() |
|
|
| topic_dropdown.change(fn=update_topic_content, inputs=topic_dropdown, outputs=[topic_content_dropdown, topic_content_state, topic_code_state]) |
| topic_content_dropdown.change(fn=update_reports, inputs=[topic_content_dropdown, topic_content_state], |
| outputs=[titles_state, topic_content_code_state, report_dropdown]) |
| report_dropdown.change(fn=update_topic_params_and_link, inputs=[report_dropdown, titles_state], |
| outputs=[topic_params_box, link_output, kwargs_box]) |
| |
| def run_get_csp_data(lang: str = 'en', topic_params_str: str = '[]', query_kwargs_str: str = '{}') -> List[Dict]: |
| """Get statistics from Official Statistics Portal of Latvia (CSP or Centrālā statistikas pārvalde). |
| |
| Args: |
| lang (str): Language. Default value 'en'. |
| topic_params_str (str): string representation of a list that is needed for data extraction. Arguments in the list should be in the following order: topic code, topic content code, topic sub-content code, report ID.\ |
| These codes you can get from the function get_titles. |
| query_kwargs_str (str): string representation of a dictionary - keyword arguments for query configuration.\ |
| Possible query argument names and their possible values can be obtained using the function get_query_values. |
| Returns: |
| list: The list of the dictionaries, where dictionary's key 'key' contains query parameters and key 'values' contains values. |
| Examples: |
| >>> # First get topic code |
| ... get_topics('Iedzīvotāji') |
| {'Iedzīvotāji': 'POP'} |
| >>> # Then use this code to get topic contents |
| ... print(get_topic_content('POP')) |
| {'Iedzīvotāju skaits un raksturojošie rādītāji': 'IR', 'Dzimstība': 'ID', 'Mirstība': 'IM', 'Nāves cēloņi': 'NC'... |
| >>> # Then use this content code to extract report titles |
| ... print(get_titles('IR')) |
| {'Iedzīvotāju skaits gada sākumā, tā izmaiņas un dabiskās kustības galvenie rādītāji 1920 - 2021': ['POP', 'IR', 'IRS010'],... |
| >>> # Use all these previous codes to select data for corresponding report |
| ... report = ['POP', 'IR', 'IRE', 'IRE010'] |
| >>> # Get 'code' and 'values' for filtering data |
| ... query_args = get_query_values(report) |
| >>> print(query_args) |
| {'ETHNICITY': {'TOTAL': 'Pavisam', 'E_LAT': 'Latvieši',...}, 'ContentsCode': {'IRE010': 'Skaits'}, 'TIME': {'1935': '1935', ..., '2025': '2025'}} |
| >>> # Get final result |
| ... data = run_get_csp_data( |
| ... lang='en', |
| ... topic_params_str = str(report), |
| ... query_kwargs_str = '{'ETHNICITY': ['E_LAT'], 'TIME': ['2024', '2025']}' |
| ... ) |
| >>> print(data[0]) |
| {'key': ['E_LAT', '2024'], 'values': ['1186337']} |
| """ |
| try: |
| topic_params = ast.literal_eval(topic_params_str) |
| query_kwargs = ast.literal_eval(query_kwargs_str) |
| if not isinstance(topic_params, list) or not isinstance(query_kwargs, dict): |
| raise ValueError("Input format error") |
| result = get_csp_data(lang=lang, topic_params=topic_params, **query_kwargs) |
| return result |
| except Exception as e: |
| return {"error": str(e)} |
|
|
| gr.api(get_topics, api_name="get_topic_name_and_id") |
| gr.api(get_topic_content, api_name="get_topic_content_name_and_id") |
| gr.api(get_titles, api_name="get_report_titles") |
| gr.api(get_query_values, api_name="get_query_values") |
|
|
| run_button.click(fn=run_get_csp_data, inputs=[lang, topic_params_box, kwargs_box], outputs=output) |
|
|
| if __name__ == "__main__": |
| demo.launch(mcp_server=True) |
|
|