Add 2025 Gatineau budget

Author: jclarkin

data/gatineau/README.md

@@ -0,0 +1,65 @@
+# Gatineau Budget Data
+
+Scripts and data for extracting and processing Gatineau's municipal budget data.
+
+## Setup
+
+```bash
+python3 -m venv venv
+source venv/bin/activate
+pip install -r data/gatineau/scripts/requirements.txt
+```
+
+## Usage
+
+### Full Workflow
+
+```bash
+export OPENAI_API_KEY='your-key-here'
+python3 data/gatineau/scripts/processor.py \
+  --year 2025 \
+  --pdf-url "https://www.gatineau.ca/..." \
+  --extract
+```
+
+### Step-by-Step
+
+**Download PDF:**
+
+```bash
+python3 data/gatineau/scripts/processor.py --year 2025 --pdf-url "https://..."
+```
+
+**Extract Data (requires API key):**
+
+```bash
+export OPENAI_API_KEY='your-key-here'
+python3 data/gatineau/scripts/processor.py --year 2025 --extract
+```
+
+**Or Extract Manually:**
+
+1. Open `raw/Gatineau Budget <year>.pdf`
+2. Use prompt from `llm_prompt.txt` with your LLM
+3. Save output to `extracted/<year>/llm_extracted.md`
+
+**Convert to JSON:**
+
+```bash
+python3 data/gatineau/scripts/processor.py --year 2025
+```
+
+## Output
+
+- `sankey.json` - Hierarchical budget data for visualization
+- `summary.json` - Metadata, metrics, and ministry breakdowns
+
+## Data Format
+
+Extracted markdown must include:
+
+- `## Key Metrics – <year>` section (population, employees, debt, property tax)
+- `## Revenues – <year>` section with nested bullets
+- `## Expenses – <year>` section with nested bullets
+
+All amounts in thousands of dollars (`k$`). See `llm_prompt.txt` for full specification.

data/gatineau/extracted/2025/llm_extracted.md

@@ -0,0 +1,83 @@
+## Key Metrics – 2025
+
+- Debt interest – **63 111 k$**
+- Property tax revenue – **748 509 k$**
+
+---
+
+## Revenues – 2025
+
+- **Taxes**
+  - Sur la valeur foncière – **587 687 k$**
+  - Compensations tenant lieu de taxes – **83 618 k$**
+  - Sur une autre base – **77 205 k$**
+  - **Total Taxes – 748 509 k$**
+    (Check: 587 687 + 83 618 + 77 205 = 748 510 → rounding difference of 1 k$)
+
+- **Subventions gouvernementales**
+  - Subventions – **42 285 k$**
+  - **Total Subventions – 42 285 k$**
+
+- **Services rendus**
+  - Tarification – **15 766 k$**
+  - Revenus recouvrables – **871 k$**
+  - Ententes – **806 k$**
+  - Autres services rendus – **606 k$**
+  - Activités commerciales – **31 k$**
+  - **Total Services rendus – 18 080 k$**
+    (Check: 15 766 + 871 + 806 + 606 + 31 = 18 080)
+
+- **Imposition de droits**
+  - Droits de mutation immobilière – **35 100 k$**
+  - Licences et permis – **5 183 k$**
+  - **Total Imposition de droits – 40 283 k$**
+    (Check: 35 100 + 5 183 = 40 283)
+
+- **Autres revenus**
+  - Revenus d’intérêts – **20 685 k$**
+  - Amendes et pénalités – **15 769 k$**
+  - Affectations – **1 349 k$**
+  - Autres revenus – **0 k$**
+  - **Total Autres revenus – 37 803 k$**
+    (Check: 20 685 + 15 769 + 1 349 + 0 = 37 803)
+
+- **Total Revenus – 886 960 k$**
+  (Check: 748 509 + 42 285 + 18 080 + 40 283 + 37 803 = 886 960)
+
+---
+
+## Expenses – 2025
+
+- **Rémunération**
+  - Salaires – **300 259 k$**
+  - Charges sociales – **82 633 k$**
+  - **Total Rémunération – 382 892 k$**
+    (Check: 300 259 + 82 633 = 382 892)
+
+- **Biens et services**
+  - Services professionnels, techniques et autres – **66 531 k$**
+  - Fourniture de service et biens non durables – **51 266 k$**
+  - Entretien et réparation – **28 567 k$**
+  - Location – **19 221 k$**
+  - Transport et communication – **4 625 k$**
+  - **Total Biens et services – 170 209 k$**
+    (Check: sum = 170 210 → rounding difference of 1 k$)
+
+- **Service de la dette**
+  - Paiements du service de la dette – **63 111 k$**
+  - **Total Service de la dette – 63 111 k$**
+
+- **Contribution à des organismes**
+  - Transport en commun – **95 340 k$**
+  - Contributions diverses à des organismes – **20 244 k$**
+  - **Total Contribution à des organismes – 115 584 k$**
+    (Check: 95 340 + 20 244 = 115 584)
+
+- **Autres dépenses**
+  - Immobilisations payées comptant – **117 701 k$**
+  - Autres dépenses – **37 463 k$**
+  - **Total Autres dépenses – 155 164 k$**
+    (Check: 117 701 + 37 463 = 155 164)
+
+- **Total Dépenses – 886 960 k$**
+  (Check: 382 892 + 170 209 + 63 111 + 115 584 + 155 164 = 886 960)

data/gatineau/llm_prompt.txt

@@ -0,0 +1,43 @@
+The attached document is the City of Gatineau budget. Extract the information for the target financial year only (ignore other years). Output MUST follow this structure:
+
+## Key Metrics – <year>
+* Population – **<number of residents>** (or omit if not available)
+* Total employees – **<number of employees>** (or omit if not available)
+* Net debt – **<amount in k$>** (or omit if not available)
+* Total debt – **<amount in k$>** (or omit if not available)
+* Debt interest – **<amount in k$>** (or omit if not available)
+* Property tax revenue – **<amount in k$>** (or omit if not available)
+
+- Use plain numbers for population/employees (no commas or units).
+- Monetary values MUST be in thousands of dollars and suffixed with `k$`.
+- If a figure is not available in the source document, omit that line entirely (do not write `0`).
+
+## Revenues – <year>
+* **Category Name** (top-level category, no amount on this line)
+  * Sub-category – **123,456 k$** (sub-item with amount)
+  * Sub-category with children (sub-item that has children):
+    * Sub-sub-category – **50,000 k$**
+    * Sub-sub-category – **30,000 k$**
+    * **Total Sub-category – 80,000 k$** (total for this sub-category)
+  * Another sub-category – **40,000 k$**
+  * **Total Category Name – 120,000 k$** (total for the top-level category)
+- Format rules:
+  * Top-level categories: `* **Category Name**` (bold, no amount, no dash)
+  * Sub-items with amounts: `  * Item Name – **amount k$**` (2 spaces indent, dash separator)
+  * Sub-items with children: `  * Item Name` (2 spaces indent, no amount, no dash)
+  * Totals: `  * **Total Item Name – amount k$**` (bold, includes dash and amount)
+  * Use 2 spaces per indentation level
+  * Subchildren can have further subchildren - go as deep as the document provides
+  * Always include a total line for each category/subcategory that has children
+
+## Expenses – <year>
+* Follow the exact same format as Revenues above.
+* Preserve all levels of hierarchy from the source document.
+* Each category must have a total line if it has sub-items.
+
+Rules:
+1. Only include numbers for the requested year.
+2. Sub-bullets must never sum to more than their parent (call out rounding differences if needed).
+3. Be precise: double-check math, preserve signage, and format every monetary amount as `123,456 k$`.
+4. Do not add commentary outside the structure above.
+

data/gatineau/raw/Gatineau Budget 2025.pdf

@@ -0,0 +1,179 @@
+{
+  "city": "Gatineau",
+  "total": 0.88696,
+  "spending": 0.88696,
+  "revenue": 0.88696,
+  "budget_balance": 0.0,
+  "spending_data": {
+    "name": "Spending",
+    "children": [
+      {
+        "name": "Rémunération",
+        "children": [
+          {
+            "name": "Salaires",
+            "amount": 0.300259
+          },
+          {
+            "name": "Charges sociales",
+            "amount": 0.082633
+          }
+        ]
+      },
+      {
+        "name": "Biens et services",
+        "children": [
+          {
+            "name": "Services professionnels, techniques et autres",
+            "amount": 0.066531
+          },
+          {
+            "name": "Fourniture de service et biens non durables",
+            "amount": 0.051266
+          },
+          {
+            "name": "Entretien et réparation",
+            "amount": 0.028567
+          },
+          {
+            "name": "Location",
+            "amount": 0.019221
+          },
+          {
+            "name": "Transport et communication",
+            "amount": 0.004625
+          }
+        ]
+      },
+      {
+        "name": "Service de la dette",
+        "children": [
+          {
+            "name": "Paiements du service de la dette",
+            "amount": 0.063111
+          }
+        ]
+      },
+      {
+        "name": "Contribution à des organismes",
+        "children": [
+          {
+            "name": "Transport en commun",
+            "amount": 0.09534
+          },
+          {
+            "name": "Contributions diverses à des organismes",
+            "amount": 0.020244
+          }
+        ]
+      },
+      {
+        "name": "Autres dépenses",
+        "children": [
+          {
+            "name": "Immobilisations payées comptant",
+            "amount": 0.117701
+          },
+          {
+            "name": "Autres dépenses",
+            "amount": 0.037463
+          }
+        ]
+      }
+    ]
+  },
+  "revenue_data": {
+    "name": "Revenue",
+    "children": [
+      {
+        "name": "Taxes",
+        "children": [
+          {
+            "name": "Sur la valeur foncière",
+            "amount": 0.587687
+          },
+          {
+            "name": "Compensations tenant lieu de taxes",
+            "amount": 0.083618
+          },
+          {
+            "name": "Sur une autre base",
+            "amount": 0.077205
+          }
+        ]
+      },
+      {
+        "name": "Subventions gouvernementales",
+        "children": [
+          {
+            "name": "Subventions",
+            "amount": 0.042285
+          }
+        ]
+      },
+      {
+        "name": "Services rendus",
+        "children": [
+          {
+            "name": "Tarification",
+            "amount": 0.015766
+          },
+          {
+            "name": "Revenus recouvrables",
+            "amount": 0.000871
+          },
+          {
+            "name": "Ententes",
+            "amount": 0.000806
+          },
+          {
+            "name": "Autres services rendus",
+            "amount": 0.000606
+          },
+          {
+            "name": "Activités commerciales",
+            "amount": 3.1e-5
+          }
+        ]
+      },
+      {
+        "name": "Imposition de droits",
+        "children": [
+          {
+            "name": "Droits de mutation immobilière",
+            "amount": 0.0351
+          },
+          {
+            "name": "Licences et permis",
+            "amount": 0.005183
+          }
+        ]
+      },
+      {
+        "name": "Autres revenus",
+        "children": [
+          {
+            "name": "Revenus d’intérêts",
+            "amount": 0.020685
+          },
+          {
+            "name": "Amendes et pénalités",
+            "amount": 0.015769
+          },
+          {
+            "name": "Affectations",
+            "amount": 0.001349
+          },
+          {
+            "name": "Autres revenus",
+            "amount": 0.0
+          }
+        ]
+      }
+    ]
+  },
+  "population": null,
+  "per_capita_spending": null,
+  "property_tax_per_capita": null,
+  "property_tax_revenue": 0.748509
+}