chdb 2.2.0b1__cp310-cp310-macosx_10_15_x86_64.whl → 3.0.1__cp310-cp310-macosx_10_15_x86_64.whl

chdb/__init__.py

@@ -1,5 +1,6 @@
 import sys
 import os
+import threading
 
 
 class ChdbError(Exception):
@@ -18,7 +19,7 @@ _process_result_format_funs = {
 # UDF script path will be f"{g_udf_path}/{func_name}.py"
 g_udf_path = ""
 
-chdb_version = ('2', '2', '0b1')
+chdb_version = ('3', '0', '1')
 if sys.version_info[:2] >= (3, 7):
     # get the path of the current file
     current_path = os.path.dirname(os.path.abspath(__file__))
@@ -29,7 +30,9 @@ if sys.version_info[:2] >= (3, 7):
     from . import _chdb  # noqa
 
     os.chdir(cwd)
-    engine_version = str(_chdb.query("SELECT version();", "CSV").bytes())[3:-4]
+    conn = _chdb.connect()
+    engine_version = str(conn.query("SELECT version();", "CSV").bytes())[3:-4]
+    conn.close()
 else:
     raise NotImplementedError("Python 3.6 or lower version is not supported")
 
@@ -64,18 +67,44 @@ def to_df(r):
     return t.to_pandas(use_threads=True)
 
 
+# global connection lock, for multi-threading use of legacy chdb.query()
+g_conn_lock = threading.Lock()
+
+
 # wrap _chdb functions
 def query(sql, output_format="CSV", path="", udf_path=""):
     global g_udf_path
     if udf_path != "":
         g_udf_path = udf_path
+    conn_str = ""
+    if path == "":
+        conn_str = ":memory:"
+    else:
+        conn_str = f"{path}"
+    if g_udf_path != "":
+        if "?" in conn_str:
+            conn_str = f"{conn_str}&udf_path={g_udf_path}"
+        else:
+            conn_str = f"{conn_str}?udf_path={g_udf_path}"
+    if output_format == "Debug":
+        output_format = "CSV"
+        if "?" in conn_str:
+            conn_str = f"{conn_str}&verbose&log-level=test"
+        else:
+            conn_str = f"{conn_str}?verbose&log-level=test"
+
     lower_output_format = output_format.lower()
     result_func = _process_result_format_funs.get(lower_output_format, lambda x: x)
     if lower_output_format in _arrow_format:
         output_format = "Arrow"
-    res = _chdb.query(sql, output_format, path=path, udf_path=g_udf_path)
-    if res.has_error():
-        raise ChdbError(res.error_message())
+
+    with g_conn_lock:
+        conn = _chdb.connect(conn_str)
+        res = conn.query(sql, output_format)
+        if res.has_error():
+            conn.close()
+            raise ChdbError(res.error_message())
+        conn.close()
     return result_func(res)
 
 

chdb/dbapi/converters.py

@@ -59,6 +59,7 @@ def escape_float(value, mapping=None):
 
 _escape_table = [chr(x) for x in range(128)]
 _escape_table[ord("'")] = u"''"
+_escape_table[ord("\\")] = "\\\\"
 
 
 def _escape_unicode(value, mapping=None):

chdb/session/state.py

@@ -1,42 +1,107 @@
 import tempfile
 import shutil
+import warnings
 
-from chdb import query
+import chdb
+from ..state import sqlitelike as chdb_stateful
+
+
+g_session = None
+g_session_path = None
 
 
 class Session:
     """
-    Session will keep the state of query. All DDL and DML state will be kept in a dir.
-    Dir path could be passed in as an argument. If not, a temporary dir will be created.
+    Session will keep the state of query.
+    If path is None, it will create a temporary directory and use it as the database path
+    and the temporary directory will be removed when the session is closed.
+    You can also pass in a path to create a database at that path where will keep your data.
+
+    You can also use a connection string to pass in the path and other parameters.
+    Examples:
+        - ":memory:" (for in-memory database)
+        - "test.db" (for relative path)
+        - "file:test.db" (same as above)
+        - "/path/to/test.db" (for absolute path)
+        - "file:/path/to/test.db" (same as above)
+        - "file:test.db?param1=value1&param2=value2" (for relative path with query params)
+        - "file::memory:?verbose&log-level=test" (for in-memory database with query params)
+        - "///path/to/test.db?param1=value1&param2=value2" (for absolute path)
 
-    If path is not specified, the temporary dir will be deleted when the Session object is deleted.
-    Otherwise path will be kept.
+    Connection string args handling:
+        Connection string can contain query params like "file:test.db?param1=value1&param2=value2"
+        "param1=value1" will be passed to ClickHouse engine as start up args.
 
-    Note: The default database is "_local" and the default engine is "Memory" which means all data
-    will be stored in memory. If you want to store data in disk, you should create another database.
+        For more details, see `clickhouse local --help --verbose`
+        Some special args handling:
+        - "mode=ro" would be "--readonly=1" for clickhouse (read-only mode)
+
+    Important:
+        - There can be only one session at a time. If you want to create a new session, you need to close the existing one.
+        - Creating a new session will close the existing one.
     """
 
     def __init__(self, path=None):
-        if path is None:
+        global g_session, g_session_path
+        if g_session is not None:
+            warnings.warn(
+                "There is already an active session. Creating a new session will close the existing one. "
+                "It is recommended to close the existing session before creating a new one. "
+                f"Closing the existing session {g_session_path}"
+            )
+            g_session.close()
+            g_session_path = None
+        if path is None or ":memory:" in path:
             self._cleanup = True
             self._path = tempfile.mkdtemp()
         else:
             self._cleanup = False
             self._path = path
+        if chdb.g_udf_path != "":
+            self._udf_path = chdb.g_udf_path
+            # add udf_path to conn_str here.
+            # - the `user_scripts_path` will be the value of `udf_path`
+            # - the `user_defined_executable_functions_config` will be `user_scripts_path/*.xml`
+            # Both of them will be added to the conn_str in the Connection class
+            if "?" in self._path:
+                self._conn_str = f"{self._path}&udf_path={self._udf_path}"
+            else:
+                self._conn_str = f"{self._path}?udf_path={self._udf_path}"
+        else:
+            self._udf_path = ""
+            self._conn_str = f"{self._path}"
+        self._conn = chdb_stateful.Connection(self._conn_str)
+        g_session = self
+        g_session_path = self._path
 
     def __del__(self):
-        if self._cleanup:
-            self.cleanup()
+        self.close()
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, traceback):
-        self.cleanup()
+        self.close()
+
+    def close(self):
+        if self._cleanup:
+            self.cleanup()
+        if self._conn is not None:
+            self._conn.close()
+            self._conn = None
+        global g_session, g_session_path
+        g_session = None
+        g_session_path = None
 
     def cleanup(self):
         try:
+            if self._conn is not None:
+                self._conn.close()
+                self._conn = None
             shutil.rmtree(self._path)
+            global g_session, g_session_path
+            g_session = None
+            g_session_path = None
         except:  # noqa
             pass
 
@@ -44,7 +109,14 @@ class Session:
         """
         Execute a query.
         """
-        return query(sql, fmt, path=self._path, udf_path=udf_path)
+        if fmt == "Debug":
+            warnings.warn(
+                """Debug format is not supported in Session.query
+Please try use parameters in connection string instead:
+Eg: conn = connect(f"db_path?verbose&log-level=test")"""
+            )
+            fmt = "CSV"
+        return self._conn.query(sql, fmt)
 
     # alias sql = query
     sql = query

chdb/state/sqlitelike.py

@@ -21,7 +21,7 @@ class Connection:
         self._cursor = Cursor(self._conn)
         return self._cursor
 
-    def query(self, query: str, format: str = "ArrowStream") -> Any:
+    def query(self, query: str, format: str = "CSV") -> Any:
         return self._conn.query(query, format)
 
     def close(self) -> None:
@@ -109,13 +109,14 @@ def connect(connection_string: str = ":memory:") -> Connection:
 
     Args:
         connection_string (str, optional): Connection string. Defaults to ":memory:".
-        Aslo support file path like:
+        Also support file path like:
           - ":memory:" (for in-memory database)
           - "test.db" (for relative path)
           - "file:test.db" (same as above)
           - "/path/to/test.db" (for absolute path)
           - "file:/path/to/test.db" (same as above)
           - "file:test.db?param1=value1&param2=value2" (for relative path with query params)
+          - "file::memory:?verbose&log-level=test" (for in-memory database with query params)
           - "///path/to/test.db?param1=value1&param2=value2" (for absolute path)
 
         Connection string args handling:

{chdb-2.2.0b1.dist-info → chdb-3.0.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: chdb
-Version: 2.2.0b1
+Version: 3.0.1
 Summary: chDB is an in-process SQL OLAP Engine powered by ClickHouse
 Home-page: https://github.com/chdb-io/chdb
 Author: auxten
@@ -30,6 +30,8 @@ Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
 License-File: LICENSE.txt
 Requires-Dist: pyarrow>=13.0.0
 Requires-Dist: pandas>=2.0.0
+Dynamic: requires-dist
+Dynamic: requires-python
 
 <div align="center">
    <a href="https://clickhouse.com/blog/chdb-joins-clickhouse-family">📢 chDB joins the ClickHouse family 🐍+🚀</a>
@@ -66,7 +68,7 @@ Requires-Dist: pandas>=2.0.0
 
 ## Arch
 <div align="center">
-  <img src="https://github.com/chdb-io/chdb/raw/main/docs/_static/arch-chdb2.png" width="450">
+  <img src="https://github.com/chdb-io/chdb/raw/main/docs/_static/arch-chdb3.png" width="450">
 </div>
 
 ## Get Started

{chdb-2.2.0b1.dist-info → chdb-3.0.1.dist-info}/RECORD

@@ -1,28 +1,28 @@
-chdb/__init__.py,sha256=li1sCjhLx7HmRBbWtekahee1iQtJl5p69oGYIaVVplk,2994
+chdb/__init__.py,sha256=6nRIwbUg4kHRY8E8EQzDv2wT-IyR0-tsr3Mu3n_Mi9Q,3762
 chdb/__main__.py,sha256=xNNtDY38d973YM5dlxiIazcqqKhXJSpNb7JflyyrXGE,1185
-chdb/_chdb.cpython-310-darwin.so,sha256=s_5Oly4JVYMcgp7akK_HHg0BA48RYpPdAPo7ZSgHLr8,421932420
+chdb/_chdb.cpython-310-darwin.so,sha256=dmUiv-V1v02Cw0SDC-XYe-fKxd8UPS-c0hUcNpJj_4c,422085000
 chdb/rwabc.py,sha256=tbiwCrXirfrfx46wCJxS64yvFe6pVWIPGdSuvrAL5Ys,2102
 chdb/dataframe/__init__.py,sha256=1_mrZZiJwqBTnH_P8_FCbbYXIWWY5sxnaFpe3-tDLF4,680
 chdb/dataframe/query.py,sha256=ggvE8A5vtabFg9gSTp99S7LCrnIEwbWtb-PtJVT8Ct0,12759
 chdb/dbapi/__init__.py,sha256=aaNhxXNBC1ZkFr260cbGR8msOinTp0VoNTT_j8AXGUc,2205
 chdb/dbapi/connections.py,sha256=4RBO0h-B149xEicE8cXSSJl9wpXa4FQMY_4SghgEvCw,2762
-chdb/dbapi/converters.py,sha256=qS9k0Kzo_vDQxnFtsJ_3pLjlTrBK09SfdWcXdxzIrtI,7413
+chdb/dbapi/converters.py,sha256=0SDqgixUTCz0LtWke_HHzgF1lFJhpsQrR_-ky3b-JRY,7447
 chdb/dbapi/cursors.py,sha256=OXF36raoyI3MIC5SCQ5IvnCtbOnppga4Q1IKOt2EIsk,7920
 chdb/dbapi/err.py,sha256=kUI9-A8LNqBoMoo4jh2NFsLCOLoPEwh9YIuz_qMoLoM,2017
 chdb/dbapi/times.py,sha256=_qXgDaYwsHntvpIKSKXp1rrYIgtq6Z9pLyLnO2XNoL0,360
 chdb/dbapi/constants/FIELD_TYPE.py,sha256=ytFzgAnGmb9hvdsBlnK68qdZv_a6jYFIXT6VSAb60z8,370
 chdb/dbapi/constants/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 chdb/session/__init__.py,sha256=fCUROZ5L1-92o2lcASiWJpFu-80-kDoSrNfouLEmLg8,50
-chdb/session/state.py,sha256=Hc9lJGLZ97Un7ElBbeWeCZoc9p_m1k_ojQbzJMPBPKg,1341
+chdb/session/state.py,sha256=nx9KlqZyPTHAflToXCJVRBUSMjJFvyh6x2akP7Gc7h0,4360
 chdb/state/__init__.py,sha256=RVUIWDqDi7gte4Os7Mz1wPXFyFpdHT_p1klJC7QtluI,55
-chdb/state/sqlitelike.py,sha256=RSVJW5M3hH_TUBcXiH7zna21WTSGbqkOccyEiFFd8PI,4425
+chdb/state/sqlitelike.py,sha256=-2bQRs6WDREr90pe8UtaYbQK-BWSkko1Ma8cLSzLUf4,4511
 chdb/udf/__init__.py,sha256=qSMaPEre7w1pYz8uJ-iZtuu8wYOUNRcI_8UNuaOymGE,80
 chdb/udf/udf.py,sha256=z0A1RmyZrx55bykpvvS-LpVt1lMrQOexjvU5zxCdCSA,3935
 chdb/utils/__init__.py,sha256=tXRcwBRGW2YQNBZWV4Mitw5QlCu_qlSRCjllw15XHbs,171
 chdb/utils/trace.py,sha256=W-pvDoKlnzq6H_7FiWjr5_teN40UNE4E5--zbUrjOIc,2511
 chdb/utils/types.py,sha256=MGLFIjoDvu7Uc2Wy8EDY60jjue66HmMPxbhrujjrZxQ,7530
-chdb-2.2.0b1.dist-info/LICENSE.txt,sha256=isYVtNCO5910aj6e9bJJ6kQceivkLqsMlFSNYwzGGKI,11366
-chdb-2.2.0b1.dist-info/METADATA,sha256=cjvQmdOiS1pll_E2ohhXF2DVb7wdbVndY00xLuCnVi8,19444
-chdb-2.2.0b1.dist-info/WHEEL,sha256=mzY66v81iH1SQD4BkoxzJ6tEjPh3oKBwQCxVJDNXy4I,111
-chdb-2.2.0b1.dist-info/top_level.txt,sha256=se0Jj0A2-ijfMW51hIjiuNyDJPqy5xJU1G8a_IEdllI,11
-chdb-2.2.0b1.dist-info/RECORD,,
+chdb-3.0.1.dist-info/LICENSE.txt,sha256=isYVtNCO5910aj6e9bJJ6kQceivkLqsMlFSNYwzGGKI,11366
+chdb-3.0.1.dist-info/METADATA,sha256=k9gmduudI-PNBH5QCOle-dKL6ascC9StyUnX4vz6gFs,19490
+chdb-3.0.1.dist-info/WHEEL,sha256=HQquLNNRUomuzWueBNrqINFman-CGVesKQI-ZGwXyWQ,111
+chdb-3.0.1.dist-info/top_level.txt,sha256=se0Jj0A2-ijfMW51hIjiuNyDJPqy5xJU1G8a_IEdllI,11
+chdb-3.0.1.dist-info/RECORD,,

{chdb-2.2.0b1.dist-info → chdb-3.0.1.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.6.0)
+Generator: setuptools (75.8.0)
 Root-Is-Purelib: false
 Tag: cp310-cp310-macosx_10_15_x86_64