ai-microcore 4.0.0.dev1__tar.gz → 4.0.0.dev3__tar.gz

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ai-microcore
-Version: 4.0.0.dev1
+Version: 4.0.0.dev3
 Summary: # Minimalistic Foundation for AI Applications
 Keywords: llm,large language models,ai,similarity search,ai search,gpt,openai
 Author-email: Vitalii Stepanenko <mail@vitalii.in>
@@ -132,6 +132,24 @@ See [transformers installation](https://huggingface.co/docs/transformers/install
     <br>💡 <small>Setting `USE_DOT_ENV` to `false` disables reading configuration files.</small>
 3.  OS environment variables have the lowest priority.
 
+### Vector Databases
+
+Vector database functions are available via `microcore.texts`.
+Default vector database is [Chroma](https://www.trychroma.com/).
+In order to use vector database functions, you need to install the `chromadb` package:
+```bash
+pip install chromadb
+```
+By default, MicroCore will use ChromaDB PersistentClient (if corresponding package is installed).
+Alternatively, you can run Chroma as separate service and configure MicroCore to use HttpClient:
+
+```python
+from microcore import configure
+configure(
+    EMBEDDING_DB_HOST = 'localhost',
+    EMBEDDING_DB_PORT = 8000,
+)
+```
 
 ## 🌟 Core Functions
 

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/README.md

@@ -107,6 +107,24 @@ See [transformers installation](https://huggingface.co/docs/transformers/install
     <br>💡 <small>Setting `USE_DOT_ENV` to `false` disables reading configuration files.</small>
 3.  OS environment variables have the lowest priority.
 
+### Vector Databases
+
+Vector database functions are available via `microcore.texts`.
+Default vector database is [Chroma](https://www.trychroma.com/).
+In order to use vector database functions, you need to install the `chromadb` package:
+```bash
+pip install chromadb
+```
+By default, MicroCore will use ChromaDB PersistentClient (if corresponding package is installed).
+Alternatively, you can run Chroma as separate service and configure MicroCore to use HttpClient:
+
+```python
+from microcore import configure
+configure(
+    EMBEDDING_DB_HOST = 'localhost',
+    EMBEDDING_DB_PORT = 8000,
+)
+```
 
 ## 🌟 Core Functions
 

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/microcore/__init__.py

@@ -161,4 +161,4 @@ __all__ = [
     # "wrappers",
 ]
 
-__version__ = "4.0.0-dev1"
+__version__ = "4.0.0-dev3"

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/microcore/embedding_db/chromadb.py

@@ -1,3 +1,4 @@
+import logging
 from dataclasses import dataclass
 import uuid
 
@@ -17,6 +18,11 @@ class ChromaEmbeddingDB(AbstractEmbeddingDB):
 
     def __post_init__(self):
         if self.config.EMBEDDING_DB_HOST:
+            logging.info(
+                "Connecting to ChromaDB at %s:%s",
+                self.config.EMBEDDING_DB_HOST,
+                self.config.EMBEDDING_DB_PORT
+            )
             self.client = chromadb.HttpClient(
                 host=self.config.EMBEDDING_DB_HOST,
                 port=self.config.EMBEDDING_DB_PORT or 8000,

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/microcore/file_storage.py

@@ -20,6 +20,9 @@ class Storage:
 
     custom_path: str = field(default="")
 
+    def __call__(self, custom_path: str):
+        return Storage(custom_path)
+
     @property
     def path(self) -> Path:
         return Path(str(self.custom_path) or config().STORAGE_PATH)

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/microcore/logging.py

@@ -41,6 +41,20 @@ def _format_request_log_str(prompt, **kwargs) -> str:
         )
         if out.endswith("\n"):
             out = out[:-1]
+    if LoggingConfig.STRIP_REQUEST_LINES:
+        start_lines, end_lines = LoggingConfig.STRIP_REQUEST_LINES
+        max_lines = start_lines + end_lines
+        lines = out.split("\n")
+        if len(lines) > max_lines:
+            out = "\n".join(
+                lines[:start_lines]
+                + [
+                    f"{LoggingConfig.INDENT}{Fore.YELLOW}"
+                    f"...(output was truncated)..."
+                    f"{LoggingConfig.PROMPT_COLOR}"
+                ]
+                + (lines[-end_lines:] if end_lines else [])
+            )
     return out
 
 
@@ -72,6 +86,7 @@ class LoggingConfig:
     OUTPUT_METHOD: callable = print
     REQUEST_FORMATTER: callable = _format_request_log_str
     RESPONSE_FORMATTER: callable = _format_response_log_str
+    STRIP_REQUEST_LINES: tuple[int, int] | None = [40, 15]
 
 
 def _log_request(prompt, **kwargs):

{ai_microcore-4.0.0.dev1 → ai_microcore-4.0.0.dev3}/microcore/utils.py

@@ -369,3 +369,93 @@ def resolve_callable(
     except (ImportError, AttributeError, AssertionError, ValueError) as e:
         raise ValueError(f"Can't resolve callable by name '{fn}', {e}") from e
     return fn
+
+
+def levenshtein(a: str, b: str) -> int:
+    """Compute the Levenshtein edit distance between two strings.
+
+    The **Levenshtein distance** is the minimum number of single‑character
+    edits (insertions, deletions, or substitutions) required to transform one
+    string into the other.
+
+    This implementation uses the classic Wagner–Fischer dynamic‑programming
+    algorithm and stores only a single row of the DP matrix at any time,
+    reducing memory usage to be linear in the length of the shorter string.
+
+    Args:
+        a (str): First input string.
+        b (str): Second input string.
+
+    Returns:
+        int: Non‑negative integer representing the edit distance. A value of
+        ``0`` means the strings are identical.
+
+    Complexity:
+        * **Time** ``O(ab)``
+        * **Space** ``O(min(a, b))``
+
+    Examples:
+        >>> levenshtein("kitten", "sitting")
+        3
+        >>> levenshtein("graph", "giraffe")
+        4
+    """
+    if a == b:
+        return 0
+    # Ensure a is the shorter string to reduce memory
+    if len(a) > len(b):
+        a, b = b, a
+    previous = list(range(len(a) + 1))
+    for i, ch_b in enumerate(b, start=1):
+        current = [i]
+        for j, ch_a in enumerate(a, start=1):
+            cost = 0 if ch_a == ch_b else 1
+            current.append(
+                min(
+                    current[-1] + 1,       # insertion
+                    previous[j] + 1,       # deletion
+                    previous[j - 1] + cost # substitution
+                )
+            )
+        previous = current
+    return previous[-1]
+
+
+def most_similar(
+    needle: str,
+    haystack: list[str],
+    distance_func: callable = levenshtein,
+    case_sensitive: bool = False,
+) -> tuple[str, int]:
+    """
+    Find the most similar string from a list of strings using the
+    specified distance function.
+
+    Args:
+        needle (str): The word to compare against.
+        haystack (list[str]): A list of words to compare with.
+        distance_func (callable): The distance function to use for comparison.
+            Defaults to levenshtein.
+        case_sensitive (bool): If True, the comparison is case-sensitive.
+
+    Returns:
+        tuple[str, int]: A tuple containing the most similar word and its distance
+            from the given word.
+
+    Raises:
+        ValueError: If haystack is empty.
+    """
+    if not haystack:
+        raise ValueError("Haystack cannot be empty")
+
+    min_dist = float('inf')
+    most_similar_word = None
+    a = needle if case_sensitive else needle.lower()
+    for word in haystack:
+        b = word if case_sensitive else word.lower()
+        dist = distance_func(a, b)
+        if dist < min_dist:
+            min_dist = dist
+            most_similar_word = word
+
+    return most_similar_word, min_dist