feat: add number of indexed indicators to available datasets tool output #191

configurations/clients/sample/tools.yaml

@@ -70,6 +70,7 @@ tools:
       fake_call:
         tool_call_id: *available_datasets_call_id
       version: full
+      include_indicator_count: true
 
   - type: dataset_structure
     <<: *shared_settings

statgpt/app/chains/datasets_meta/available_datasets_tool.py

@@ -27,9 +27,19 @@ async def _arun(self, inputs: dict) -> tuple[str, ToolArtifact]:
         versioned_datasets = await data_service.list_available_datasets(auth_context)
         datasets = [ds.data for ds in versioned_datasets]
 
+        indicator_counts: dict[str, int] | None = None
+        if self._tool_config.details.include_indicator_count:
+            indicator_counts = await data_service.get_indicator_counts(
+                auth_context, versioned_datasets
+            )
+
         formatter = DatasetsListFormatter(self._dataset_formatter_config, auth_context=auth_context)
         response = await formatter.format(
-            datasets, sort_by_name=True, add_stats=True, group_by_provider=True
+            datasets,
+            sort_by_name=True,
+            add_stats=True,
+            group_by_provider=True,
+            indicator_counts=indicator_counts,
         )
 
         target = ChainParameters.get_target(inputs)

statgpt/app/services/chat_facade.py

@@ -411,6 +411,21 @@ async def _load_datasets(self, auth_context: AuthContext) -> list[VersionedDataS
     async def list_available_datasets(self, auth_context: AuthContext) -> list[VersionedDataSet]:
         return await self._load_datasets(auth_context)
 
+    async def get_indicator_counts(
+        self, auth_context: AuthContext, versioned_datasets: list[VersionedDataSet]
+    ) -> dict[str, int]:
+        """Get indicator count per dataset `entity_id`."""
+        vector_store = await self._get_indicators_vector_store(auth_context)
+        version_to_entity = {
+            ds.version.version_data_id: ds.data.entity_id for ds in versioned_datasets
+        }
+        counts_by_version = await vector_store.get_size_per_version(set(version_to_entity.keys()))
+        return {
+            version_to_entity[vid]: count
+            for vid, count in counts_by_version.items()
+            if vid in version_to_entity
+        }
+
     async def get_dataset_hierarchy(self, auth_context: AuthContext) -> DatasetHierarchy | None:
         """Get first available dataset hierarchy from the channel data sources."""
 

statgpt/app/utils/formatters/dataset_base.py

@@ -58,5 +58,5 @@ async def _get_dataset_update_at(self, dataset: DataSet) -> str:
         return last_updated
 
     @abstractmethod
-    async def format(self, dataset: DataSet) -> str:
+    async def format(self, dataset: DataSet, indicator_count: int | None = None) -> str:
         pass

statgpt/app/utils/formatters/dataset_detailed.py

@@ -93,9 +93,9 @@ def _append_attributes(self, dataset: DataSet, result: list[str]) -> None:
                         f'{attribute_details_tabs}- {self._("Description")}: {attr.description}'
                     )
 
-    async def format(self, dataset: DataSet) -> str:
+    async def format(self, dataset: DataSet, indicator_count: int | None = None) -> str:
         result: list[str] = []
-        await self._append_basic_info(dataset, result)
+        await self._append_basic_info(dataset, result, indicator_count=indicator_count)
         self._append_dimensions(dataset, result)
         self._append_attributes(dataset, result)
         return "\n".join(result)

statgpt/app/utils/formatters/dataset_simple.py

@@ -6,7 +6,9 @@
 
 class SimpleDatasetFormatter(BaseDatasetFormatter):
 
-    async def _append_basic_info(self, dataset: DataSet, result: list[str]) -> None:
+    async def _append_basic_info(
+        self, dataset: DataSet, result: list[str], indicator_count: int | None = None
+    ) -> None:
         if self.config.include_name:
             name_str = f'**{dataset.name}**' if self.config.highlight_name_in_bold else dataset.name
             if self.config.official_dataset_label and dataset.config.is_official:
@@ -50,7 +52,10 @@ async def _append_basic_info(self, dataset: DataSet, result: list[str]) -> None:
             ).format(citation)
             result.append(formatted_citation)
 
-    async def format(self, dataset: DataSet) -> str:
+        if indicator_count is not None:
+            result.append(f'{item_tabs}* {self._("Number of indicators")}: {indicator_count}')
+
+    async def format(self, dataset: DataSet, indicator_count: int | None = None) -> str:
         result: list[str] = []
-        await self._append_basic_info(dataset, result)
+        await self._append_basic_info(dataset, result, indicator_count=indicator_count)
         return "\n".join(result)

statgpt/app/utils/formatters/datasets_list_formatter.py

@@ -34,6 +34,7 @@ async def format(
         sort_by_name: bool = False,
         add_stats: bool = False,
         group_by_provider: bool = False,
+        indicator_counts: dict[str, int] | None = None,
     ) -> str:
         if sort_by_id and sort_by_name:
             raise ValueError(self._("Cannot sort by both id and name."))
@@ -50,7 +51,8 @@ async def format(
 
         # Format each dataset
         for dataset in iterable:
-            entry = await self._formatter.format(dataset)
+            count = indicator_counts.get(dataset.entity_id) if indicator_counts else None
+            entry = await self._formatter.format(dataset, indicator_count=count)
             provider = dataset.config.citation.provider if dataset.config.citation else None
             dataset_entries[provider].append(entry)
 
@@ -80,6 +82,10 @@ async def format(
                 # Count unique providers
                 providers = [p for p in dataset_entries.keys() if p is not None]
                 stats_header += f'\n{self._("Total providers")}: {len(providers)}'
+            if indicator_counts is not None:
+                stats_header += (
+                    f'\n{self._("Total number of indicators")}: {sum(indicator_counts.values())}'
+                )
             result = f'{stats_header}\n\n{datasets_list}'
         else:
             result = datasets_list

statgpt/app/utils/formatters/locales/dataset.pot

@@ -8,7 +8,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: PACKAGE VERSION\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2025-11-05 17:19+0200\n"
+"POT-Creation-Date: 2026-03-06 14:48+0200\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language-Team: LANGUAGE <LL@li.org>\n"
@@ -17,23 +17,27 @@ msgstr ""
 "Content-Type: text/plain; charset=CHARSET\n"
 "Content-Transfer-Encoding: 8bit\n"
 
-#: dataset_simple.py:16
+#: dataset_simple.py:18
 msgid "[Official]"
 msgstr ""
 
-#: dataset_simple.py:28 dataset_query.py:125 dataset_query.py:247
+#: dataset_simple.py:30 dataset_query.py:125 dataset_query.py:247
 #: dataset_availablity_query.py:105
 msgid "ID"
 msgstr ""
 
-#: dataset_simple.py:32
+#: dataset_simple.py:34
 msgid "Internal ID"
 msgstr ""
 
-#: dataset_simple.py:41 dataset_detailed.py:59 dataset_detailed.py:93
+#: dataset_simple.py:43 dataset_detailed.py:59 dataset_detailed.py:93
 msgid "Description"
 msgstr ""
 
+#: dataset_simple.py:56
+msgid "Number of indicators"
+msgstr ""
+
 #: dataset_detailed.py:21 dataset_detailed.py:31
 msgid "Total"
 msgstr ""
@@ -63,40 +67,44 @@ msgstr ""
 msgid "Attributes"
 msgstr ""
 
-#: datasets_list_formatter.py:39
+#: datasets_list_formatter.py:40
 msgid "Cannot sort by both id and name."
 msgstr ""
 
-#: datasets_list_formatter.py:62 datasets_list_formatter.py:64 citation.py:28
+#: datasets_list_formatter.py:64 datasets_list_formatter.py:66 citation.py:28
 msgid "Provider"
 msgstr ""
 
-#: datasets_list_formatter.py:64 datasets_list_formatter.py:110
+#: datasets_list_formatter.py:66 datasets_list_formatter.py:116
 msgid "Unknown"
 msgstr ""
 
-#: datasets_list_formatter.py:67
+#: datasets_list_formatter.py:69
 msgid "Total datasets from this provider"
 msgstr ""
 
-#: datasets_list_formatter.py:78 datasets_list_formatter.py:100
-#: datasets_list_formatter.py:105
+#: datasets_list_formatter.py:80 datasets_list_formatter.py:106
+#: datasets_list_formatter.py:111
 msgid "Total datasets"
 msgstr ""
 
-#: datasets_list_formatter.py:82
+#: datasets_list_formatter.py:84
 msgid "Total providers"
 msgstr ""
 
-#: datasets_list_formatter.py:101
+#: datasets_list_formatter.py:87
+msgid "Total number of indicators"
+msgstr ""
+
+#: datasets_list_formatter.py:107
 msgid "Official"
 msgstr ""
 
-#: datasets_list_formatter.py:102
+#: datasets_list_formatter.py:108
 msgid "Unofficial"
 msgstr ""
 
-#: datasets_list_formatter.py:114
+#: datasets_list_formatter.py:120
 msgid "By provider"
 msgstr ""
 

statgpt/app/utils/formatters/locales/en/LC_MESSAGES/dataset.po

@@ -6,7 +6,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: statgpt 1.0\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2025-11-05 17:19+0200\n"
+"POT-Creation-Date: 2026-03-06 14:48+0200\n"
 "PO-Revision-Date: 2024-01-01 00:00+0000\n"
 "Last-Translator: \n"
 "Language-Team: English\n"
@@ -15,24 +15,28 @@ msgstr ""
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
 
-#: dataset_simple.py:16
+#: dataset_simple.py:18
 msgid "[Official]"
 msgstr "[Official]"
 
 # Basic labels
-#: dataset_simple.py:28 dataset_query.py:125 dataset_query.py:247
+#: dataset_simple.py:30 dataset_query.py:125 dataset_query.py:247
 #: dataset_availablity_query.py:105
 msgid "ID"
 msgstr "ID"
 
-#: dataset_simple.py:32
+#: dataset_simple.py:34
 msgid "Internal ID"
 msgstr "Internal ID"
 
-#: dataset_simple.py:41 dataset_detailed.py:59 dataset_detailed.py:93
+#: dataset_simple.py:43 dataset_detailed.py:59 dataset_detailed.py:93
 msgid "Description"
 msgstr "Description"
 
+#: dataset_simple.py:56
+msgid "Number of indicators"
+msgstr "Number of indicators"
+
 #: dataset_detailed.py:21 dataset_detailed.py:31
 msgid "Total"
 msgstr "Total"
@@ -63,40 +67,44 @@ msgid "Attributes"
 msgstr "Attributes"
 
 # List formatter strings
-#: datasets_list_formatter.py:39
+#: datasets_list_formatter.py:40
 msgid "Cannot sort by both id and name."
 msgstr "Cannot sort by both id and name."
 
-#: datasets_list_formatter.py:62 datasets_list_formatter.py:64 citation.py:28
+#: datasets_list_formatter.py:64 datasets_list_formatter.py:66 citation.py:28
 msgid "Provider"
 msgstr "Provider"
 
-#: datasets_list_formatter.py:64 datasets_list_formatter.py:110
+#: datasets_list_formatter.py:66 datasets_list_formatter.py:116
 msgid "Unknown"
 msgstr "Unknown"
 
-#: datasets_list_formatter.py:67
+#: datasets_list_formatter.py:69
 msgid "Total datasets from this provider"
 msgstr "Total datasets from this provider"
 
-#: datasets_list_formatter.py:78 datasets_list_formatter.py:100
-#: datasets_list_formatter.py:105
+#: datasets_list_formatter.py:80 datasets_list_formatter.py:106
+#: datasets_list_formatter.py:111
 msgid "Total datasets"
 msgstr "Total datasets"
 
-#: datasets_list_formatter.py:82
+#: datasets_list_formatter.py:84
 msgid "Total providers"
 msgstr "Total providers"
 
-#: datasets_list_formatter.py:101
+#: datasets_list_formatter.py:87
+msgid "Total number of indicators"
+msgstr "Total number of indicators"
+
+#: datasets_list_formatter.py:107
 msgid "Official"
 msgstr "Official"
 
-#: datasets_list_formatter.py:102
+#: datasets_list_formatter.py:108
 msgid "Unofficial"
 msgstr "Unofficial"
 
-#: datasets_list_formatter.py:114
+#: datasets_list_formatter.py:120
 msgid "By provider"
 msgstr "By provider"