wolframalpha: move to its own plugin

config.json.example

@@ -200,6 +200,14 @@
       "ambiguous_response": "It is ambiguous.",
       "empty_response": "I can't find it."
     },
+    "wolframalpha": {
+      "commands": ["wolfram"],
+      "aliases": {
+        "wolfram": ["wolframalpha", "compute"]
+      },
+      "api_key": "",
+      "max_pods": 1
+    },
     "yell": {
       "commands": ["yell"],
       "target_word": "to",

edmond/plugins/unknown_command.py

@@ -2,29 +2,12 @@ from edmond.plugin import Plugin
 
 
 class UnknownCommandPlugin(Plugin):
-    """Handle unknown command by sending it to WolframAlpha."""
-
-    REQUIRED_CONFIGS = ["api_key", "max_pods"]
-    MAX_LENGTH = 256
-    CUT_MARK = "..."
 
     def __init__(self, bot):
         super().__init__(bot)
         self.priority = -6
-        self._client = None
-
-    @property
-    def client(self):
-        if self._client is None and self.has_api_key():
-            self._client = wolframalpha.Client(self.config["api_key"])
-        return self._client
-
-    def has_api_key(self):
-        return self.config["api_key"] != ""
 
     def on_pubmsg(self, event):
-        if not self.has_api_key:
-            return False
         message = self.should_read_message(event.arguments[0])
         if not message:
             return False
@@ -33,51 +16,5 @@ class UnknownCommandPlugin(Plugin):
             return False
 
         query = " ".join(words[:-1])
-        self.process_query(query, event.target)
+        self.bot.say(event.target, f"UnknownCommand: {query}")
         return True
-
-    def process_query(self, query, target):
-        self.bot.log_d(f"Processing '{query}' with WolframAlpha.")
-        try:
-            response = self.client.query(query)
-        except Exception as exc:  # unstable lib
-            self.bot.log_w(f"wolframalpha exception: {exc}")
-            self.signal_failure(target)
-            return
-
-        if not response["@success"]:
-            self.bot.log_d("Call to WA succeeded but response is an error.")
-            self.signal_failure(target)
-            return
-
-        inputs = []
-        answers = []
-        num_pods = 0
-        for pod in response.pods:
-            for subpod in pod.subpods:
-                self.bot.log_d(f"WolframAlpha subpod: {subpod}")
-                if pod["@id"] == "Input":
-                    inputs.append(self.sanitize_text(subpod.plaintext or ""))
-                else:
-                    answers.append(self.sanitize_text(subpod.plaintext or ""))
-                    num_pods += 1
-                if num_pods >= self.config["max_pods"]:
-                    break
-            if num_pods >= self.config["max_pods"]:
-                break
-
-        input_text = ", ".join(inputs)
-        answer_text = ", ".join(answers)
-        if input_text:
-            reply = input_text + " -- " + answer_text
-        else:
-            reply = answer_text
-        if len(reply) > self.MAX_LENGTH - len(self.CUT_MARK):
-            reply = (
-                reply[: self.MAX_LENGTH - len(self.CUT_MARK)] + self.CUT_MARK
-            )
-        self.bot.say(target, reply)
-
-    @staticmethod
-    def sanitize_text(text):
-        return text.replace("\n", ", ")

edmond/plugins/wolfram_alpha.py

@@ -0,0 +1,79 @@
+import wolframalpha
+
+from edmond.plugin import Plugin
+
+
+class WolframAlphaPlugin(Plugin):
+    """Handle unknown command by sending it to WolframAlpha."""
+
+    REQUIRED_CONFIGS = ["commands", "api_key", "max_pods"]
+    MAX_LENGTH = 256
+    CUT_MARK = "..."
+
+    def __init__(self, bot):
+        super().__init__(bot)
+        self.priority = -6
+        self._client = None
+
+    @property
+    def client(self):
+        if self._client is None and self.has_api_key():
+            self._client = wolframalpha.Client(self.config["api_key"])
+        return self._client
+
+    def on_welcome(self, _):
+        if not self.config["api_key"]:
+            self.bot.log_w("API key unavailable.")
+            self.is_ready = False
+
+    def on_pubmsg(self, event):
+        if not self.should_handle_command(event.arguments[0]):
+            return False
+        self.process_query(self.command.content, event.target)
+        return True
+
+    def process_query(self, query, target):
+        self.bot.log_d(f"Processing '{query}' with WolframAlpha.")
+        try:
+            response = self.client.query(query)
+        except Exception as exc:  # unstable lib
+            self.bot.log_w(f"wolframalpha exception: {exc}")
+            self.signal_failure(target)
+            return
+
+        if not response["@success"]:
+            self.bot.log_d("Call to WA succeeded but response is an error.")
+            self.signal_failure(target)
+            return
+
+        inputs = []
+        answers = []
+        num_pods = 0
+        for pod in response.pods:
+            for subpod in pod.subpods:
+                self.bot.log_d(f"WolframAlpha subpod: {subpod}")
+                if pod["@id"] == "Input":
+                    inputs.append(self.sanitize_text(subpod.plaintext or ""))
+                else:
+                    answers.append(self.sanitize_text(subpod.plaintext or ""))
+                    num_pods += 1
+                if num_pods >= self.config["max_pods"]:
+                    break
+            if num_pods >= self.config["max_pods"]:
+                break
+
+        input_text = ", ".join(inputs)
+        answer_text = ", ".join(answers)
+        if input_text:
+            reply = input_text + " -- " + answer_text
+        else:
+            reply = answer_text
+        if len(reply) > self.MAX_LENGTH - len(self.CUT_MARK):
+            reply = (
+                reply[: self.MAX_LENGTH - len(self.CUT_MARK)] + self.CUT_MARK
+            )
+        self.bot.say(target, reply)
+
+    @staticmethod
+    def sanitize_text(text):
+        return text.replace("\n", ", ")