jupyter-duckdb 1.4.100__tar.gz → 1.4.105__tar.gz

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jupyter-duckdb
-Version: 1.4.100
+Version: 1.4.105
 Summary: a basic wrapper kernel for DuckDB
 Home-page: https://github.com/erictroebs/jupyter-duckdb
 Author: Eric Tröbs
@@ -45,6 +45,7 @@ There are some magic commands that make teaching easier with this kernel.
     - [Ship Tests With Your Notebook](#ship-tests-with-your-notebooks)
     - [Relational Algebra](#relational-algebra)
     - [Domain Calculus](#domain-calculus)
+    - [Automated Parser Selection](#automated-parser-selection)
 
 ## Setup
 
@@ -281,6 +282,8 @@ The supported operations are:
 The optional flag `ANALYZE` can be used to add an execution diagram to the
 output.
 
+You can also add comments to queries using `--` or `/* */`, just like in SQL.
+
 The Dockerfile also installs the Jupyter Lab plugin
 [jupyter-ra-extension](https://pypi.org/project/jupyter-ra-extension/). It adds
 the symbols mentioned above and some other supported symbols to the toolbar for
@@ -304,4 +307,3 @@ executed cells.
 If the magic command `%AUTO_PARSER` is added to a cell, a parser is
 automatically selected. If `%GUESS_PARSER` is executed, the parser is
 automatically selected for all subsequent cells.
-

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/README.md

@@ -17,6 +17,7 @@ There are some magic commands that make teaching easier with this kernel.
     - [Ship Tests With Your Notebook](#ship-tests-with-your-notebooks)
     - [Relational Algebra](#relational-algebra)
     - [Domain Calculus](#domain-calculus)
+    - [Automated Parser Selection](#automated-parser-selection)
 
 ## Setup
 
@@ -253,6 +254,8 @@ The supported operations are:
 The optional flag `ANALYZE` can be used to add an execution diagram to the
 output.
 
+You can also add comments to queries using `--` or `/* */`, just like in SQL.
+
 The Dockerfile also installs the Jupyter Lab plugin
 [jupyter-ra-extension](https://pypi.org/project/jupyter-ra-extension/). It adds
 the symbols mentioned above and some other supported symbols to the toolbar for
@@ -276,4 +279,3 @@ executed cells.
 If the magic command `%AUTO_PARSER` is added to a cell, a parser is
 automatically selected. If `%GUESS_PARSER` is executed, the parser is
 automatically selected for all subsequent cells.
-

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/setup.py

@@ -1,7 +1,7 @@
 import os
 
 # configuration
-PACKAGE_VERSION = '1.4.100'
+PACKAGE_VERSION = '1.4.105'
 DUCKDB_VERSION = '1.4.1'
 
 DEPENDENCIES = [

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/kernel.py

@@ -13,6 +13,7 @@ from .db import Connection, DatabaseError, Table
 from .db.error import *
 from .magics import *
 from .parser import RAParser, DCParser, ParserError
+from .parser.util.QuerySplitter import split_queries, get_last_query
 from .util.ResultSetComparator import ResultSetComparator
 from .util.SQL import SQL_KEYWORDS
 from .util.TestError import TestError
@@ -27,9 +28,11 @@ class DuckDBKernel(Kernel):
     implementation_version = '1.0'
     banner = 'DuckDB Kernel'
     language_info = {
-        'name': 'duckdb',
-        'mimetype': 'application/sql',
+        'name': 'sql',
         'file_extension': '.sql',
+        'mimetype': 'text/x-sql',
+        'codemirror_mode': 'sql',
+        'pygments_lexer': 'sql',
     }
 
     def __init__(self, **kwargs):
@@ -166,7 +169,9 @@ class DuckDBKernel(Kernel):
         # print result if not silent
         if not silent:
             # print EXPLAIN queries as raw text if using DuckDB
-            if query.strip().startswith('EXPLAIN') and state.db.plain_explain():
+            last_query = get_last_query(query, remove_comments=True).strip()
+
+            if last_query.startswith('EXPLAIN') and state.db.plain_explain():
                 for ekey, evalue in rows:
                     html = f'<b>{ekey}</b><br><pre>{evalue}</pre>'
                     break
@@ -273,8 +278,7 @@ class DuckDBKernel(Kernel):
 
                     # You can only execute one statement at a time using SQLite.
                     if not state.db.multiple_statements_per_query():
-                        statements = re.split(r';\r?\n', content)
-                        for statement in statements:
+                        for statement in split_queries(content):
                             try:
                                 state.db.execute(statement)
                             except EmptyResultError:

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/parser/RAParser.py

@@ -2,6 +2,7 @@ from .LogicParser import LogicParser
 from .ParserError import RAParserError
 from .elements import *
 from .tokenizer import *
+from .util.QuerySplitter import get_last_query
 
 
 # Instead of multiple nested loops, a tree with rotation can
@@ -10,6 +11,10 @@ from .tokenizer import *
 class RAParser:
     @staticmethod
     def parse_query(query: str) -> RAElement:
+        # remove comments from query
+        query = get_last_query(query, split_at=None, remove_comments=True)
+
+        # parse query
         initial_token = Token(query)
         return RAParser.parse_tokens(initial_token, depth=0)
 

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/parser/elements/RAElement.py

@@ -36,10 +36,13 @@ class RAElement:
         # if all columns are from the same relation we can skip the relation name
         if len(set(c.table for c in columns)) == 1:
             column_names = ', '.join(f'{c.current_name} AS "{c.name}"' for c in columns)
+            order_names = ', '.join(f'"{c.name}" ASC' for c in columns)
         else:
             column_names = ', '.join(f'{c.current_name} AS "{c.full_name}"' for c in columns)
+            order_names = ', '.join(f'"{c.full_name}" ASC' for c in columns)
 
-        return f'SELECT {column_names} FROM ({sql}) {self._name()}'
+        # create sql
+        return f'SELECT {column_names} FROM ({sql}) {self._name()} ORDER BY {order_names}'
 
     def to_sql_with_count(self, tables: Dict[str, Table]) -> str:
         sql, _ = self.to_sql(tables)

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/parser/elements/binary/ConditionalSet.py

@@ -347,7 +347,13 @@ class ConditionalSet:
             sql_join_filters += f' AND {join_filter}'
 
         sql_condition = condition.to_sql(joined_columns) if condition is not None else '1=1'
-        sql_query = f'SELECT DISTINCT {sql_select} FROM {sql_tables} WHERE ({sql_join_filters}) AND ({sql_condition})'
+
+        if self.projection == ('*',):
+            sql_order = ', '.join(f'{rc.name} ASC' for rcl in rcls for rc in rcl)
+        else:
+            sql_order = ', '.join(f'{col} ASC' for col in self.projection)
+
+        sql_query = f'SELECT DISTINCT {sql_select} FROM {sql_tables} WHERE ({sql_join_filters}) AND ({sql_condition}) ORDER BY {sql_order}'
 
         # Create a mapping from intermediate column names to constant values.
         column_name_mapping = {

jupyter_duckdb-1.4.105/src/duckdb_kernel/parser/util/QuerySplitter.py

@@ -0,0 +1,87 @@
+from typing import Iterator
+
+
+def split_queries(query: str, split_at: str | None = ';', remove_comments: bool = False) -> Iterator[str]:
+    quotes = '\'"`'
+
+    escaped = False
+    in_quotes = None
+    in_singleline_comment = False
+    in_multiline_comment = False
+
+    previous = None
+    current_query = []
+
+    for symbol in query:
+        keep_symbol = True
+
+        # escaped symbol
+        if escaped:
+            escaped = False
+
+        # backslash (escape)
+        elif symbol == '\\':
+            escaped = True
+
+        # if in quotes
+        elif in_quotes is not None:
+            if symbol == in_quotes:
+                in_quotes = False
+
+        # if in single line comment
+        elif in_singleline_comment:
+            if symbol == '\n':
+                in_singleline_comment = False
+            elif remove_comments:
+                keep_symbol = False
+
+        # if in multiline comment
+        elif in_multiline_comment:
+            if previous == '*' and symbol == '/':
+                in_multiline_comment = False
+
+            if remove_comments:
+                keep_symbol = False
+
+        # start of quotes
+        elif symbol in quotes:
+            in_quotes = symbol
+
+        # start of single line comment
+        elif previous == '-' and symbol == '-':
+            in_singleline_comment = True
+
+            if remove_comments:
+                keep_symbol = False
+                current_query.pop()
+
+        # start of multiline comment
+        elif previous == '/' and symbol == '*':
+            in_multiline_comment = True
+
+            if remove_comments:
+                keep_symbol = False
+                current_query.pop()
+
+        # semicolon
+        elif split_at is not None and symbol == split_at:
+            yield ''.join(current_query)
+
+            current_query = []
+            keep_symbol = False
+
+        # store symbol
+        if keep_symbol:
+            current_query.append(symbol)
+
+        previous = symbol
+
+    # yield remaining symbols
+    yield ''.join(current_query)
+
+
+def get_last_query(query: str, split_at: str | None = ';', remove_comments: bool = False) -> str:
+    for query in split_queries(query, split_at, remove_comments):
+        pass
+
+    return query

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/tests/__init__.py

@@ -56,7 +56,7 @@ class Connection:
         sql = root.to_sql_with_renamed_columns(self.tables)
         cols, rows = self.execute_sql_return_cols(sql)
 
-        return cols, sorted(rows, key=lambda t: tuple(-1 if x is None else x for x in t))
+        return cols, rows  # sorted(rows, key=lambda t: tuple(-1 if x is None else x for x in t))
 
     def execute_ra(self, root: RAElement) -> List:
         _, rows = self.execute_ra_return_cols(root)
@@ -66,7 +66,7 @@ class Connection:
         sql, cnm = root.to_sql_with_renamed_columns(self.tables)
         cols, rows = self.execute_sql_return_cols(sql)
 
-        return [cnm.get(c, c) for c in cols], sorted(rows)
+        return [cnm.get(c, c) for c in cols], rows  # sorted(rows)
 
     def execute_dc(self, root: ConditionalSet) -> List:
         _, rows = self.execute_dc_return_cols(root)

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/duckdb_kernel/tests/test_ra.py

@@ -79,6 +79,29 @@ def test_case_insensitivity():
             ]
 
 
+def test_comments():
+    for query in (
+            # single line
+            'Shows -- x Users\n x Seasons',
+            'Shows x Seasons -- x Users',
+            'Shows x Seasons--',
+            'Shows x Seasons--\n',
+            # multi line
+            'Shows /* x Users */ x Seasons',
+            'Shows /* x Users */\n x Seasons',
+            'Shows /* x Users\n */ x Seasons',
+            'Shows x Seasons/**/',
+            'Shows x Seasons/*\n*/',
+            'Shows x Seasons\n/**/',
+            'Shows x Seasons/* x Users'
+    ):
+        root = RAParser.parse_query(query)
+
+        assert isinstance(root, BinaryOperators.Cross)
+        assert isinstance(root.left, RAOperand) and root.left.name == 'Shows'
+        assert isinstance(root.right, RAOperand) and root.right.name == 'Seasons'
+
+
 def test_binary_operator_cross():
     for query in (
             r'Shows x Seasons',

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/jupyter_duckdb.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jupyter-duckdb
-Version: 1.4.100
+Version: 1.4.105
 Summary: a basic wrapper kernel for DuckDB
 Home-page: https://github.com/erictroebs/jupyter-duckdb
 Author: Eric Tröbs
@@ -45,6 +45,7 @@ There are some magic commands that make teaching easier with this kernel.
     - [Ship Tests With Your Notebook](#ship-tests-with-your-notebooks)
     - [Relational Algebra](#relational-algebra)
     - [Domain Calculus](#domain-calculus)
+    - [Automated Parser Selection](#automated-parser-selection)
 
 ## Setup
 
@@ -281,6 +282,8 @@ The supported operations are:
 The optional flag `ANALYZE` can be used to add an execution diagram to the
 output.
 
+You can also add comments to queries using `--` or `/* */`, just like in SQL.
+
 The Dockerfile also installs the Jupyter Lab plugin
 [jupyter-ra-extension](https://pypi.org/project/jupyter-ra-extension/). It adds
 the symbols mentioned above and some other supported symbols to the toolbar for
@@ -304,4 +307,3 @@ executed cells.
 If the magic command `%AUTO_PARSER` is added to a cell, a parser is
 automatically selected. If `%GUESS_PARSER` is executed, the parser is
 automatically selected for all subsequent cells.
-

{jupyter_duckdb-1.4.100 → jupyter_duckdb-1.4.105}/src/jupyter_duckdb.egg-info/SOURCES.txt

@@ -73,6 +73,7 @@ src/duckdb_kernel/parser/elements/unary/__init__.py
 src/duckdb_kernel/parser/tokenizer/Token.py
 src/duckdb_kernel/parser/tokenizer/Tokenizer.py
 src/duckdb_kernel/parser/tokenizer/__init__.py
+src/duckdb_kernel/parser/util/QuerySplitter.py
 src/duckdb_kernel/parser/util/RenamableColumn.py
 src/duckdb_kernel/parser/util/RenamableColumnList.py
 src/duckdb_kernel/parser/util/__init__.py