curl-programming-lang 1.3.0__tar.gz → 1.4.0__tar.gz

{curl_programming_lang-1.3.0/curl_programming_lang.egg-info → curl_programming_lang-1.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: curl-programming-lang
-Version: 1.3.0
+Version: 1.4.0
 Summary: Curl is an open-source programming language built on Python technology
 Author: Ritvik Gautam
 License: Apache-2.0

curl_programming_lang-1.4.0/ai_module.py

@@ -0,0 +1,159 @@
+import os
+import json
+import urllib.request
+import urllib.error
+
+_DEFAULT_SYSTEM = (
+    "You are Curl-Bot, an AI assistant built into the Curl programming language. "
+    "Curl is an open-source language created by Ritvik Gautam that runs on Python. "
+    "You are helpful, concise, and friendly."
+)
+
+# Auto-compact after this many user+assistant message pairs in history
+_COMPACT_THRESHOLD = 20
+
+
+class CurlAIModule:
+    """
+    AI standard library for Curl.
+
+    Configuration via environment variables:
+      CURL_AI_KEY or OPENAI_API_KEY  — API key (not needed for local models)
+      CURL_AI_BASE_URL               — base URL (default: https://api.openai.com/v1)
+                                       set to http://localhost:11434/v1 for Ollama
+      CURL_AI_MODEL                  — model name (default: gpt-4o-mini)
+
+    Usage in Curl:
+      import{"ai", ai}\
+      ai.context{"You are a pirate who only speaks in rhymes."}\
+      var{answer, ai.ask{"What is the capital of France?"}}\
+      pcType{var{answer}}\
+    """
+
+    def __init__(self):
+        self._system = _DEFAULT_SYSTEM
+        self._history = []  # list of {"role": ..., "content": ...}
+
+    def context(self, prompt):
+        """Set a persistent system prompt (persona/context) for all subsequent AI calls."""
+        self._system = str(prompt)
+        self._history = []  # reset history when context changes
+        return ""
+
+    def reset(self, _=""):
+        """Clear conversation history (start a fresh chat while keeping context)."""
+        self._history = []
+        return ""
+
+    def _raw_request(self, messages):
+        """Send a messages list to the API and return the reply string."""
+        api_key = os.environ.get("CURL_AI_KEY") or os.environ.get("OPENAI_API_KEY", "")
+        base_url = os.environ.get("CURL_AI_BASE_URL", "https://api.openai.com/v1").rstrip("/")
+        model = os.environ.get("CURL_AI_MODEL", "gpt-4o-mini")
+
+        if not api_key and "openai.com" in base_url:
+            raise RuntimeError(
+                "No API key found. Set CURL_AI_KEY (or OPENAI_API_KEY) in your environment.\n"
+                "For a free local model, install Ollama (https://ollama.com), run a model, then set:\n"
+                "  export CURL_AI_BASE_URL=http://localhost:11434/v1\n"
+                "  export CURL_AI_MODEL=llama3.2"
+            )
+
+        payload = json.dumps({
+            "model": model,
+            "messages": messages,
+            "temperature": 0.7,
+        }).encode()
+
+        headers = {"Content-Type": "application/json"}
+        if api_key:
+            headers["Authorization"] = f"Bearer {api_key}"
+
+        req = urllib.request.Request(
+            f"{base_url}/chat/completions",
+            data=payload,
+            headers=headers,
+        )
+
+        try:
+            with urllib.request.urlopen(req, timeout=60) as resp:
+                result = json.loads(resp.read())
+                return result["choices"][0]["message"]["content"].strip()
+        except urllib.error.HTTPError as e:
+            body = e.read().decode(errors="ignore")
+            raise RuntimeError(f"AI request failed ({e.code}): {body}")
+        except urllib.error.URLError as e:
+            raise RuntimeError(f"AI connection failed: {e.reason}")
+
+    def _compact_history(self):
+        """Summarize old history into a single assistant message to stay within limits."""
+        summary_messages = [
+            {"role": "system", "content": self._system},
+            {
+                "role": "user",
+                "content": (
+                    "Summarize the following conversation history into 3-5 bullet points "
+                    "so it can be used as context going forward. Be concise.\n\n"
+                    + "\n".join(
+                        f"{m['role'].upper()}: {m['content']}" for m in self._history
+                    )
+                ),
+            },
+        ]
+        summary = self._raw_request(summary_messages)
+        self._history = [{"role": "assistant", "content": f"[Previous conversation summary]\n{summary}"}]
+
+    def _request(self, prompt):
+        """Send prompt with history, auto-compact when history is too long."""
+        if len(self._history) >= _COMPACT_THRESHOLD:
+            self._compact_history()
+
+        self._history.append({"role": "user", "content": str(prompt)})
+
+        messages = [{"role": "system", "content": self._system}] + self._history
+        reply = self._raw_request(messages)
+
+        self._history.append({"role": "assistant", "content": reply})
+        return reply
+
+    def ask(self, prompt):
+        """Send a prompt (with conversation history) and return the response."""
+        return self._request(prompt)
+
+    def summarize(self, text):
+        """Summarize text in 2-3 sentences (no history tracking)."""
+        messages = [
+            {"role": "system", "content": self._system},
+            {"role": "user", "content": f"Summarize the following in 2-3 sentences:\n\n{text}"},
+        ]
+        return self._raw_request(messages)
+
+    def analyze(self, text):
+        """Analyze text and return key insights (no history tracking)."""
+        messages = [
+            {"role": "system", "content": self._system},
+            {"role": "user", "content": f"Analyze the following and provide key insights:\n\n{text}"},
+        ]
+        return self._raw_request(messages)
+
+    def sentiment(self, text):
+        """Return the sentiment of text: positive, negative, or neutral (no history tracking)."""
+        messages = [
+            {"role": "system", "content": self._system},
+            {
+                "role": "user",
+                "content": (
+                    "What is the sentiment of this text? "
+                    "Reply with only one word: positive, negative, or neutral.\n\n" + text
+                ),
+            },
+        ]
+        return self._raw_request(messages)
+
+    def translate(self, text):
+        """Translate text — include the target language in the text itself (no history tracking)."""
+        messages = [
+            {"role": "system", "content": self._system},
+            {"role": "user", "content": f"Translate the following:\n\n{text}"},
+        ]
+        return self._raw_request(messages)

{curl_programming_lang-1.3.0 → curl_programming_lang-1.4.0/curl_programming_lang.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: curl-programming-lang
-Version: 1.3.0
+Version: 1.4.0
 Summary: Curl is an open-source programming language built on Python technology
 Author: Ritvik Gautam
 License: Apache-2.0

{curl_programming_lang-1.3.0 → curl_programming_lang-1.4.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "curl-programming-lang"
-version = "1.3.0"
+version = "1.4.0"
 description = "Curl is an open-source programming language built on Python technology"
 readme = "README.md"
 requires-python = ">=3.7"

curl_programming_lang-1.3.0/ai_module.py

@@ -1,83 +0,0 @@
-import os
-import json
-import urllib.request
-import urllib.error
-
-
-class CurlAIModule:
-    """
-    AI standard library for Curl.
-
-    Configuration via environment variables:
-      CURL_AI_KEY or OPENAI_API_KEY  — API key (not needed for local models)
-      CURL_AI_BASE_URL               — base URL (default: https://api.openai.com/v1)
-                                       set to http://localhost:11434/v1 for Ollama
-      CURL_AI_MODEL                  — model name (default: gpt-4o-mini)
-
-    Usage in Curl:
-      import{"ai", ai}\
-      var{answer, ai.ask{"What is the capital of France?"}}\
-      pcType{var{answer}}\
-    """
-
-    def _request(self, prompt):
-        api_key = os.environ.get("CURL_AI_KEY") or os.environ.get("OPENAI_API_KEY", "")
-        base_url = os.environ.get("CURL_AI_BASE_URL", "https://api.openai.com/v1").rstrip("/")
-        model = os.environ.get("CURL_AI_MODEL", "gpt-4o-mini")
-
-        if not api_key and "openai.com" in base_url:
-            raise RuntimeError(
-                "No API key found. Set CURL_AI_KEY (or OPENAI_API_KEY) in your environment.\n"
-                "For a free local model, install Ollama (https://ollama.com), run a model, then set:\n"
-                "  export CURL_AI_BASE_URL=http://localhost:11434/v1\n"
-                "  export CURL_AI_MODEL=llama3.2"
-            )
-
-        payload = json.dumps({
-            "model": model,
-            "messages": [{"role": "user", "content": str(prompt)}],
-            "temperature": 0.7,
-        }).encode()
-
-        headers = {"Content-Type": "application/json"}
-        if api_key:
-            headers["Authorization"] = f"Bearer {api_key}"
-
-        req = urllib.request.Request(
-            f"{base_url}/chat/completions",
-            data=payload,
-            headers=headers,
-        )
-
-        try:
-            with urllib.request.urlopen(req, timeout=60) as resp:
-                result = json.loads(resp.read())
-                return result["choices"][0]["message"]["content"].strip()
-        except urllib.error.HTTPError as e:
-            body = e.read().decode(errors="ignore")
-            raise RuntimeError(f"AI request failed ({e.code}): {body}")
-        except urllib.error.URLError as e:
-            raise RuntimeError(f"AI connection failed: {e.reason}")
-
-    def ask(self, prompt):
-        """Send a prompt and return the response as a string."""
-        return self._request(prompt)
-
-    def summarize(self, text):
-        """Summarize text in 2-3 sentences."""
-        return self._request(f"Summarize the following in 2-3 sentences:\n\n{text}")
-
-    def analyze(self, text):
-        """Analyze text and return key insights."""
-        return self._request(f"Analyze the following and provide key insights:\n\n{text}")
-
-    def sentiment(self, text):
-        """Return the sentiment of text: positive, negative, or neutral."""
-        return self._request(
-            f"What is the sentiment of this text? "
-            f"Reply with only one word: positive, negative, or neutral.\n\n{text}"
-        )
-
-    def translate(self, text):
-        """Translate text — include the target language in the text itself."""
-        return self._request(f"Translate the following:\n\n{text}")