From 59431d618a3c5933e9c09ae1b97df7f4a1581130 Mon Sep 17 00:00:00 2001
From: Jorge <jorgecardleitao@gmail.com>
Date: Sat, 10 Feb 2024 08:14:31 +0000
Subject: [PATCH] Added by country

---
 analysis.py | 106 ++++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 99 insertions(+), 7 deletions(-)

diff --git a/analysis.py b/analysis.py
index d5b0634..8534eb6 100644
--- a/analysis.py
+++ b/analysis.py
@@ -35,16 +35,19 @@ def set_size(width: float, fraction=1):
 FIG_SIZE = set_size(600)
 
 
-STATUS_PATH = "https://private-jets.fra1.digitaloceanspaces.com/leg/v1/status.json"
-LEGS_PATH = "https://private-jets.fra1.digitaloceanspaces.com/leg/v1/all.csv"
+VERSION = "v1"
+STATUS_PATH = f"https://private-jets.fra1.digitaloceanspaces.com/leg/{VERSION}/status.json"
+LEGS_PATH = f"https://private-jets.fra1.digitaloceanspaces.com/leg/{VERSION}/all.csv"
 AIRCRAFT_PATH = "https://private-jets.fra1.digitaloceanspaces.com/private_jets/all.csv"
 duckdb.sql(
     f"""
 CREATE TEMP TABLE "legs" AS (
 SELECT
     *
-FROM read_csv_auto('{LEGS_PATH}', header = true)
-WHERE date_part('year', "start") = 2023
+FROM
+    read_csv_auto('{LEGS_PATH}', header = true)
+WHERE
+    date_part('year', "start") = 2023
 )
 """
 )
@@ -90,6 +93,93 @@ def aircraft():
     fig.savefig("results/aircraft.png", dpi=300)
 
 
+def aircraft_by_country():
+    x = duckdb.sql(
+        f"""
+    SELECT
+        country,COUNT(*) AS aircraft
+    FROM read_csv_auto('{AIRCRAFT_PATH}', header = true)
+    GROUP BY country
+    ORDER BY aircraft DESC
+    """
+    ).fetchall()
+
+    t = list(range(len(x)))
+    number = [x[1] for x in x]
+    percentage = [x[1] / sum(number) * 100 for x in x]
+
+    fig, (ax1, ax2) = plt.subplots(2, 1, figsize=FIG_SIZE, sharex=True)
+
+    #fig.supylabel("Number of aircraft")
+    ax2.set_xlabel("Country (rank by number)")
+    ax1.grid(linestyle="dotted")
+    ax2.grid(linestyle="dotted")
+
+    ax1.annotate(x[0][0], xy=(t[0] + 1, number[0]))
+    for i in range(1, 5):
+        ax2.annotate(x[i][0], xy=(t[i] + 1, number[i]))
+    ax1.plot(t, number, "o")
+    ax2.plot(t, number, "o")
+    # ax1.tick_params(axis="y", labelcolor=color)
+    ax1.set_ylim(0.6 * sum(number), 1.0 * sum(number))
+    ax2.set_ylim(0, 0.04 * sum(number))
+
+    # hide the spines between ax and ax2
+    ax1.spines.bottom.set_visible(False)
+    ax2.spines.top.set_visible(False)
+    ax1.xaxis.tick_top()
+    ax1.tick_params(labeltop=False)  # don't put tick labels at the top
+    ax2.xaxis.tick_bottom()
+
+    d = 0.5  # proportion of vertical to horizontal extent of the slanted line
+    kwargs = dict(
+        marker=[(-1, -d), (1, d)],
+        markersize=12,
+        linestyle="none",
+        color="k",
+        mec="k",
+        mew=1,
+        clip_on=False,
+    )
+    ax1.plot([0, 1], [0, 0], transform=ax1.transAxes, **kwargs)
+    ax2.plot([0, 1], [1, 1], transform=ax2.transAxes, **kwargs)
+
+    ax3 = ax1.twinx()
+
+    fig.text(
+            0.03,
+            0.5,
+            "Number of aircraft",
+            ha="center",
+            va="center",
+            rotation="vertical",
+        )
+    fig.text(
+        0.97,
+        0.5,
+        "Market share (%)",
+        ha="center",
+        va="center",
+        rotation="vertical",
+    )
+
+    ax3.plot(t, percentage, "o")
+    ax3.set_ylim(60, 100)
+
+    ax4 = ax2.twinx()
+
+    ax4.plot(t, percentage, "o")
+    ax4.set_ylim(0.0, 4)
+
+    ax4.set_zorder(1)  # default zorder is 0 for ax1 and ax2
+    ax4.patch.set_visible(False)  # prevents ax1 from hiding ax2
+
+    fig.subplots_adjust(hspace=0.05)
+    fig.savefig("results/country.png", dpi=300, bbox_inches="tight")
+
+    return
+
+
 def ranked_legs():
     x = duckdb.sql(
         f"""
@@ -251,6 +341,8 @@ def per_month():
 
     fig.tight_layout()
     fig.savefig("results/timeseries_emissions.png", dpi=300)
+    print(f"{sum(emissions)} kt of CO2e")
+    print(f"{sum(emissions) / 3.0 / 1.68} Mt of CO2")
 
 
 def hours_per_model():
@@ -300,6 +392,7 @@ def hours_per_model():
 
 def base_analysis():
     aircraft()
+    aircraft_by_country()
     ranked_legs()
     ranked_hours()
     histogram_hours()
@@ -312,12 +405,11 @@ def base_analysis():
 def distribution():
     x = duckdb.sql(
         f"""
-INSTALL spatial;
-LOAD spatial;
 SELECT
     epoch("end" - "start") / 60 / 60 AS "flying_time_k_hours"
     , distance
 FROM "legs"
+USING SAMPLE 10%
     """
     ).fetchall()
 
@@ -333,5 +425,5 @@ def distribution():
     plt.savefig("results/dist.png", dpi=300)
 
 
-# distribution()
+distribution()
 base_analysis()