improved graphs with histogram and details

requirements.txt

@@ -5,3 +5,5 @@ nextcord~=2.6.0
 requests~=2.31.0
 
 cx_freeze
+matplotlib~=3.8.2
+numpy~=1.26.3

tools/statistics/analyse/age.py

@@ -1,16 +1,11 @@
-from collections import Counter
-
 import matplotlib.pyplot as plt
+import numpy as np
 
 from tools.statistics import extract
 
 
-def analyse(datas: list[dict]):
-    ages_data = list(map(extract.age.extract, datas))
-
-    counter = Counter(ages_data)
-    x = list(counter.keys())
-    y = list(counter.values())
+def analyse(datas: list[dict]) -> plt.Figure:
+    x = list(map(extract.age.extract, datas))
 
     # prepare plotting
     figure: plt.Figure = plt.figure()
@@ -18,6 +13,7 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre de personne par âge")
 
     # bar chart
-    axes.bar(x, y)
+    bins = np.arange(min(x), max(x), 1)
+    axes.hist(x, bins=bins, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/completion_per_age.py

@@ -6,7 +6,7 @@ import numpy as np
 from tools.statistics import extract
 
 
-def analyse(datas: list[dict]):
+def analyse(datas: list[dict]) -> plt.Figure:
     ages_completion: dict[int, int] = defaultdict(lambda: 0)
     ages_count: dict[int, int] = defaultdict(lambda: 0)
 
@@ -34,6 +34,7 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre moyen de mission complété par âge")
 
     # bar chart
-    axes.bar(x, y)
+    bins = np.arange(min(x), max(x), 1)
+    axes.hist(x, bins, weights=y, edgecolor="black")
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/completion_per_experience.py

@@ -1,12 +1,12 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    experience_completion: dict[str, int] = {"yes": 0, "mixed": 0, "no": 0}
-    experience_count: dict[str, int] = {"yes": 0, "mixed": 0, "no": 0}
+def analyse(datas: list[dict]) -> plt.Figure:
+    experience_completion: dict[str, int] = dict.fromkeys(ressource.experience.choices, 0)
+    experience_count: dict[str, int] = dict.fromkeys(ressource.experience.choices, 0)
 
     for data in datas:
         experience = extract.experience.extract(data)
@@ -32,6 +32,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre moyen de mission complété par expérience")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/completion_per_language.py

@@ -0,0 +1,39 @@
+from collections import defaultdict
+
+import matplotlib.pyplot as plt
+import numpy as np
+
+from tools.statistics import extract
+
+
+def analyse(datas: list[dict]) -> plt.Figure:
+    languages_completion: dict[str, int] = defaultdict(lambda: 0)
+    languages_count: dict[str, int] = defaultdict(lambda: 0)
+
+    for data in datas:
+        language = extract.language.extract(data)
+        languages_count[language] += 1
+
+        for survey, survey_data in data["surveys"].items():
+            # only scan survey mission
+            if not survey.startswith("mission-"):
+                continue
+
+            if extract.mission_completed.extract(data, survey):
+                languages_completion[language] += 1
+
+    x = list(languages_completion.keys())
+    y = (
+        np.array(list(languages_completion.values()))
+        / np.array(list(languages_count.values()))
+    )
+
+    # prepare plotting
+    figure: plt.Figure = plt.figure()
+    axes = figure.add_subplot(1, 1, 1)
+    axes.set_title("Nombre moyen de mission complété par langue")
+
+    # bar chart
+    axes.bar(x, y, edgecolor='black')
+
+    return figure

tools/statistics/analyse/completion_per_mission.py

@@ -1,23 +1,12 @@
 import matplotlib.pyplot as plt
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    completions: dict[str] = {
-        "mission-language": 0,
-        "mission-price": 0,
-        "mission-community-hub": 0,
-        "mission-game-page": 0,
-        "mission-game-dlc": 0,
-        "mission-actuality-new": 0,
-        "mission-profile": 0,
-        "mission-game-discussion": 0,
-        "mission-gift-card": 0,
-        "mission-workshop": 0,
-    }
+def analyse(datas: list[dict]) -> plt.Figure:
+    completions: dict[str] = dict.fromkeys(ressource.mission.choices, 0)
 
-    # NOTE : séparé avant / après grosse mise à jour pour carte cadeau ?
+    # TODO : couleur par mission
 
     for data in datas:
         for survey in data["surveys"].keys():
@@ -37,8 +26,8 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre de personne ayant réussi par mission")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
     axes.set_xticks(x)
     axes.set_xticklabels(x, rotation=45)
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/completion_per_usage.py

@@ -1,12 +1,12 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    usage_completion: dict[str, int] = {"always": 0, "often": 0, "sometime": 0, "rarely": 0, "never": 0}
-    usage_count: dict[str, int] = {"always": 0, "often": 0, "sometime": 0, "rarely": 0, "never": 0}
+def analyse(datas: list[dict]) -> plt.Figure:
+    usage_completion: dict[str, int] = dict.fromkeys(ressource.usage.choices, 0)
+    usage_count: dict[str, int] = dict.fromkeys(ressource.usage.choices, 0)
 
     for data in datas:
         usage = next(filter(
@@ -36,6 +36,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre moyen de mission complété par niveau")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/duration_per_age.py

@@ -6,11 +6,11 @@ import numpy as np
 from tools.statistics import extract
 
 
-def analyse(datas: list[dict]):
+def analyse(datas: list[dict]) -> plt.Figure:
     ages_duration: dict[int, int] = defaultdict(lambda: 0)
     ages_count: dict[int, int] = defaultdict(lambda: 0)
 
-    # TODO: faire des tranches d'âges ?
+    # TODO: affichage en minutes ?
 
     for data in datas:
         age = extract.age.extract(data)
@@ -35,6 +35,7 @@ def analyse(datas: list[dict]):
     axes.set_title("Temps moyen passé par âge")
 
     # bar chart
-    axes.bar(x, y)
+    bins = np.arange(min(x), max(x), 1)
+    axes.hist(x, bins, weights=y, edgecolor="black")
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/duration_per_experience.py

@@ -1,12 +1,12 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    experience_duration: dict[str, int] = {"yes": 0, "mixed": 0, "no": 0}
-    experience_count: dict[str, int] = {"yes": 0, "mixed": 0, "no": 0}
+def analyse(datas: list[dict]) -> plt.Figure:
+    experience_duration: dict[str, int] = dict.fromkeys(ressource.experience.choices, 0)
+    experience_count: dict[str, int] = dict.fromkeys(ressource.experience.choices, 0)
 
     for data in datas:
         experience = extract.experience.extract(data)
@@ -32,6 +32,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Temps moyen passé par expérience")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/duration_per_language.py

@@ -0,0 +1,38 @@
+from collections import defaultdict
+
+import matplotlib.pyplot as plt
+import numpy as np
+
+from tools.statistics import extract
+
+
+def analyse(datas: list[dict]) -> plt.Figure:
+    languages_duration: dict[str, int] = defaultdict(lambda: 0)
+    languages_count: dict[str, int] = defaultdict(lambda: 0)
+
+    for data in datas:
+        language = extract.language.extract(data)
+        languages_count[language] += 1
+
+        for survey in data["surveys"].keys():
+            # only scan survey mission
+            if not survey.startswith("mission-"):
+                continue
+
+            languages_duration[language] += extract.mission_duration.extract(data, survey)
+
+    x = list(languages_duration.keys())
+    y = (
+        np.array(list(languages_duration.values()))
+        / np.array(list(languages_count.values()))
+    )
+
+    # prepare plotting
+    figure: plt.Figure = plt.figure()
+    axes = figure.add_subplot(1, 1, 1)
+    axes.set_title("Temps moyen passé par langue")
+
+    # bar chart
+    axes.bar(x, y, edgecolor='black')
+
+    return figure

tools/statistics/analyse/duration_per_mission.py

@@ -1,25 +1,14 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    durations: dict[str] = {
-        "mission-language": 0,
-        "mission-price": 0,
-        "mission-community-hub": 0,
-        "mission-game-page": 0,
-        "mission-game-dlc": 0,
-        "mission-actuality-new": 0,
-        "mission-profile": 0,
-        "mission-game-discussion": 0,
-        "mission-gift-card": 0,
-        "mission-workshop": 0,
-    }
+def analyse(datas: list[dict]) -> plt.Figure:
+    durations: dict[str] = dict.fromkeys(ressource.mission.choices, 0)
 
-    # NOTE : séparé avant / après grosse mise à jour pour carte cadeau ?
-    # NOTE : marqué en rouge la durée d'abandon ?
+    # TODO : marqué en rouge la durée d'abandon ?
+    # TODO : couleur par mission
 
     for data in datas:
         for survey in data["surveys"].keys():
@@ -38,8 +27,8 @@ def analyse(datas: list[dict]):
     axes.set_title("Temps moyen passé par test")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
     axes.set_xticks(x)
     axes.set_xticklabels(x, rotation=45)
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/duration_per_usage.py

@@ -1,14 +1,12 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    usage_completion: dict[str, int] = {"always": 0, "often": 0, "sometime": 0, "rarely": 0, "never": 0}
-    usage_count: dict[str, int] = {"always": 0, "often": 0, "sometime": 0, "rarely": 0, "never": 0}
-
-    # TODO: faire des tranches d'âges ?
+def analyse(datas: list[dict]) -> plt.Figure:
+    usage_completion: dict[str, int] = dict.fromkeys(ressource.usage.choices, 0)
+    usage_count: dict[str, int] = dict.fromkeys(ressource.usage.choices, 0)
 
     for data in datas:
         usage = extract.usage.extract(data)
@@ -33,6 +31,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Temps moyen passé par niveau")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/experience.py

@@ -2,11 +2,11 @@ from collections import Counter
 
 import matplotlib.pyplot as plt
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    experiences: dict[str, int] = {"yes": 0, "mixed": 0, "no": 0}
+def analyse(datas: list[dict]) -> plt.Figure:
+    experiences: dict[str, int] = dict.fromkeys(ressource.experience.choices, 0)
     for data in datas:
         experience = extract.experience.extract(data)
         experiences[experience] += 1
@@ -21,6 +21,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Nombre de personne par expérience")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/hardest_mission.py

@@ -1,21 +1,10 @@
 import matplotlib.pyplot as plt
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
-    missions = {
-        "mission-language": 0,
-        "mission-price": 0,
-        "mission-community-hub": 0,
-        "mission-game-page": 0,
-        "mission-game-dlc": 0,
-        "mission-actuality-new": 0,
-        "mission-profile": 0,
-        "mission-game-discussion": 0,
-        "mission-gift-card": 0,
-        "mission-workshop": 0
-    }
+def analyse(datas: list[dict]) -> plt.Figure:
+    missions = dict.fromkeys(ressource.mission.choices, 0)
 
     for data in datas:
         missions[extract.hardest_mission.extract(data)] += 1
@@ -29,8 +18,8 @@ def analyse(datas: list[dict]):
     axes.set_title("Mission la plus difficile des personnes sondées")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
     axes.set_xticks(x)
     axes.set_xticklabels(x, rotation=45)
 
-    plt.show(block=True)
+    return figure

tools/statistics/analyse/language.py

@@ -0,0 +1,23 @@
+from collections import Counter
+
+import matplotlib.pyplot as plt
+
+from tools.statistics import extract
+
+
+def analyse(datas: list[dict]) -> plt.Figure:
+    languages = list(map(extract.language.extract, datas))
+
+    counter = Counter(languages)
+    x = list(counter.keys())
+    y = list(counter.values())
+
+    # prepare plotting
+    figure: plt.Figure = plt.figure()
+    axes = figure.add_subplot(1, 1, 1)
+    axes.set_title("Langue des personnes sondées")
+
+    # bar chart
+    axes.bar(x, y, edgecolor='black')
+
+    return figure

tools/statistics/analyse/usage.py

@@ -1,12 +1,12 @@
 import matplotlib.pyplot as plt
 
-from tools.statistics import extract
+from tools.statistics import extract, ressource
 
 
-def analyse(datas: list[dict]):
+def analyse(datas: list[dict]) -> plt.Figure:
     usage_data = list(map(extract.usage.extract, datas))
 
-    usages: dict[str, int] = {"always": 0, "often": 0, "sometime": 0, "rarely": 0, "never": 0}
+    usages: dict[str, int] = dict.fromkeys(ressource.usage.choices, 0)
     for usage in usage_data:
         usages[usage] += 1
 
@@ -19,6 +19,6 @@ def analyse(datas: list[dict]):
     axes.set_title("Expérience antérieure des personnes sondées")
 
     # bar chart
-    axes.bar(x, y)
+    axes.bar(x, y, edgecolor='black')
 
-    plt.show(block=True)
+    return figure

tools/statistics/extract/__init__.py

@@ -4,3 +4,4 @@ from . import mission_duration
 from . import mission_completed
 from . import experience
 from . import hardest_mission
+from . import language

tools/statistics/extract/language.py

@@ -0,0 +1,2 @@
+def extract(data: dict) -> str:
+    return data["language"]

tools/statistics/main.py

@@ -2,30 +2,63 @@ from pathlib import Path
 
 from tools.statistics.analyse import (age, usage, completion_per_mission, duration_per_mission, completion_per_age,
                                       completion_per_usage, duration_per_age, duration_per_usage,
-                                      completion_per_experience, duration_per_experience, experience, hardest_mission)
+                                      completion_per_experience, duration_per_experience, experience, hardest_mission,
+                                      language, duration_per_language, completion_per_language)
 
 if __name__ == "__main__":
     from source.utils import compress
 
-    directory = Path(r"./sondage/")
+    # import matplotlib
+    # matplotlib.use("pgf")
+    # matplotlib.rcParams.update({
+    #     "pgf.texsystem": "pdflatex",
+    #     'font.family': 'serif',
+    #     'font.size': 11,
+    #     'text.usetex': True,
+    #     'pgf.rcfonts': False,
+    # })
 
-    # read every people survey data
-    datas = [
+    sondage_path = Path(r"./sondage/")
+    graph_path = Path(r"./graph/")
+    graph_path.mkdir(parents=True, exist_ok=True)
+
+    # read every peoples survey data
+    datas_all = [
         compress.uncompress_data(file.read_bytes())  # decompress the data
-        for file in directory.rglob("*.rsl")
+        for file in sondage_path.rglob("*.rsl")
     ]
 
-    age.analyse(datas)
-    usage.analyse(datas)
-    experience.analyse(datas)
-    hardest_mission.analyse(datas)
+    # keep only the datas before the steam new year update
+    datas_steam_version_1 = list(filter(lambda data: data["time"] < 1704409200, datas_all))
 
-    completion_per_mission.analyse(datas)
-    completion_per_age.analyse(datas)
-    completion_per_usage.analyse(datas)
-    completion_per_experience.analyse(datas)
+    # keep only the datas after the steam new year update
+    datas_steam_version_2 = list(filter(lambda data: data["time"] >= 1704409200, datas_all))
 
-    duration_per_mission.analyse(datas)
-    duration_per_age.analyse(datas)
-    duration_per_usage.analyse(datas)
-    duration_per_experience.analyse(datas)
+    # regroup all the datas
+    datasets = {
+        "all": datas_all,
+        "version_1": datas_steam_version_1,
+        "version_2": datas_steam_version_2,
+    }
+
+    for datas_name, datas in datasets.items():
+        directory = graph_path / datas_name
+        directory.mkdir(parents=True, exist_ok=True)
+
+        age.analyse(datas).savefig(directory / "age.svg")
+        usage.analyse(datas).savefig(directory / "usage.svg")
+        experience.analyse(datas).savefig(directory / "experience.svg")
+        hardest_mission.analyse(datas).savefig(directory / "hardest_mission.svg")
+        language.analyse(datas).savefig(directory / "language.svg")
+
+        completion_per_mission.analyse(datas).savefig(directory / "completion_per_mission.svg")
+        completion_per_age.analyse(datas).savefig(directory / "completion_per_age.svg")
+        completion_per_usage.analyse(datas).savefig(directory / "completion_per_usage.svg")
+        completion_per_experience.analyse(datas).savefig(directory / "completion_per_experience.svg")
+        completion_per_language.analyse(datas).savefig(directory / "completion_per_language.svg")
+
+        duration_per_mission.analyse(datas).savefig(directory / "duration_per_mission.svg")
+        duration_per_age.analyse(datas).savefig(directory / "duration_per_age.svg")
+        duration_per_usage.analyse(datas).savefig(directory / "duration_per_usage.svg")
+        duration_per_experience.analyse(datas).savefig(directory / "duration_per_experience.svg")
+        duration_per_language.analyse(datas).savefig(directory / "duration_per_language.svg")

tools/statistics/ressource/__init__.py

@@ -0,0 +1,3 @@
+from . import experience
+from . import mission
+from . import usage

tools/statistics/ressource/experience.py

@@ -0,0 +1 @@
+choices = ["yes", "mixed", "no"]

tools/statistics/ressource/mission.py

@@ -0,0 +1,12 @@
+choices = [
+    "mission-language",
+    "mission-price",
+    "mission-community-hub",
+    "mission-game-page",
+    "mission-game-dlc",
+    "mission-actuality-new",
+    "mission-profile",
+    "mission-game-discussion",
+    "mission-gift-card",
+    "mission-workshop",
+]

tools/statistics/ressource/usage.py

@@ -0,0 +1 @@
+choices = ["always", "often", "sometime", "rarely", "never"]

tools/statistics/survey/IntegerQuestion.py

@@ -1,6 +0,0 @@
-class IntegerQuestion:
-    def get_average(self, datas: list[int]):
-        return sum(datas) / len(datas)
-
-    def get_median(self, datas: list[int]):
-        return datas[len(datas) // 2]