datachain 0.3.3__py3-none-any.whl → 0.3.4__py3-none-any.whl

datachain/sql/functions/string.py

@@ -26,5 +26,17 @@ class split(GenericFunction):  # noqa: N801
     inherit_cache = True
 
 
+class regexp_replace(GenericFunction):  # noqa: N801
+    """
+    Replaces substring that match a regular expression.
+    """
+
+    type = String()
+    package = "string"
+    name = "regexp_replace"
+    inherit_cache = True
+
+
 compiler_not_implemented(length)
 compiler_not_implemented(split)
+compiler_not_implemented(regexp_replace)

datachain/sql/sqlite/base.py

@@ -1,4 +1,5 @@
 import logging
+import re
 import sqlite3
 from collections.abc import Iterable
 from datetime import MAXYEAR, MINYEAR, datetime, timezone
@@ -77,6 +78,7 @@ def setup():
     compiles(array.length, "sqlite")(compile_array_length)
     compiles(string.length, "sqlite")(compile_string_length)
     compiles(string.split, "sqlite")(compile_string_split)
+    compiles(string.regexp_replace, "sqlite")(compile_regexp_replace)
     compiles(conditional.greatest, "sqlite")(compile_greatest)
     compiles(conditional.least, "sqlite")(compile_least)
     compiles(Values, "sqlite")(compile_values)
@@ -178,9 +180,15 @@ def register_user_defined_sql_functions() -> None:
 
     _registered_function_creators["vector_functions"] = create_vector_functions
 
+    def sqlite_regexp_replace(string: str, pattern: str, replacement: str) -> str:
+        return re.sub(pattern, replacement, string)
+
     def create_string_functions(conn):
         conn.create_function("split", 2, sqlite_string_split, deterministic=True)
         conn.create_function("split", 3, sqlite_string_split, deterministic=True)
+        conn.create_function(
+            "regexp_replace", 3, sqlite_regexp_replace, deterministic=True
+        )
 
     _registered_function_creators["string_functions"] = create_string_functions
 
@@ -265,6 +273,10 @@ def path_file_ext(path):
     return func.substr(path, func.length(path) - path_file_ext_length(path) + 1)
 
 
+def compile_regexp_replace(element, compiler, **kwargs):
+    return f"regexp_replace({compiler.process(element.clauses, **kwargs)})"
+
+
 def compile_path_parent(element, compiler, **kwargs):
     return compiler.process(path_parent(*element.clauses.clauses), **kwargs)
 

{datachain-0.3.3.dist-info → datachain-0.3.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: datachain
-Version: 0.3.3
+Version: 0.3.4
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License: Apache-2.0

{datachain-0.3.3.dist-info → datachain-0.3.4.dist-info}/RECORD

@@ -88,15 +88,15 @@ datachain/sql/functions/array.py,sha256=EB7nJSncUc1PuxlHyzU2gVhF8DuXaxpGlxb5e8X2
 datachain/sql/functions/conditional.py,sha256=q7YUKfunXeEldXaxgT-p5pUTcOEVU_tcQ2BJlquTRPs,207
 datachain/sql/functions/path.py,sha256=zixpERotTFP6LZ7I4TiGtyRA8kXOoZmH1yzH9oRW0mg,1294
 datachain/sql/functions/random.py,sha256=vBwEEj98VH4LjWixUCygQ5Bz1mv1nohsCG0-ZTELlVg,271
-datachain/sql/functions/string.py,sha256=hIrF1fTvlPamDtm8UMnWDcnGfbbjCsHxZXS30U2Rzxo,651
+datachain/sql/functions/string.py,sha256=NSQIpmtQgm68hz3TFJsgHMBuo4MjBNhDSyEIC3pWkT8,916
 datachain/sql/sqlite/__init__.py,sha256=TAdJX0Bg28XdqPO-QwUVKy8rg78cgMileHvMNot7d04,166
-datachain/sql/sqlite/base.py,sha256=5nLvOv0xcOlEpfZeY3SWbI401MSGM2i29P3SRkd7TAc,12898
+datachain/sql/sqlite/base.py,sha256=WLPHBhZbXbiqPoRV1VgDrXJqku4UuvJpBhYeQ0k5rI8,13364
 datachain/sql/sqlite/types.py,sha256=yzvp0sXSEoEYXs6zaYC_2YubarQoZH-MiUNXcpuEP4s,1573
 datachain/sql/sqlite/vector.py,sha256=ncW4eu2FlJhrP_CIpsvtkUabZlQdl2D5Lgwy_cbfqR0,469
 datachain/torch/__init__.py,sha256=gIS74PoEPy4TB3X6vx9nLO0Y3sLJzsA8ckn8pRWihJM,579
-datachain-0.3.3.dist-info/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
-datachain-0.3.3.dist-info/METADATA,sha256=BDBQIVMBj7tqy0TntMooUyMlPEVgVHA4xvMESRHiF0I,16789
-datachain-0.3.3.dist-info/WHEEL,sha256=HiCZjzuy6Dw0hdX5R3LCFPDmFS4BWl8H-8W39XfmgX4,91
-datachain-0.3.3.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
-datachain-0.3.3.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
-datachain-0.3.3.dist-info/RECORD,,
+datachain-0.3.4.dist-info/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
+datachain-0.3.4.dist-info/METADATA,sha256=nV1-yJcDxoWuaM8uSwEzYCpDqSMxhxPle3EZZ98a-LA,16789
+datachain-0.3.4.dist-info/WHEEL,sha256=nCVcAvsfA9TDtwGwhYaRrlPhTLV9m-Ga6mdyDtuwK18,91
+datachain-0.3.4.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
+datachain-0.3.4.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
+datachain-0.3.4.dist-info/RECORD,,

{datachain-0.3.3.dist-info → datachain-0.3.4.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (72.2.0)
+Generator: setuptools (73.0.0)
 Root-Is-Purelib: true
 Tag: py3-none-any