Fix identifier case handling

[hab6]

Description

Removed forced conversion of identifiers to lowercase so that uppercase and mixed-case scenarios now work properly.

Also removed encoding/decoding of string to bytes. Proper handling (if needed) will be dealt with via #36.

Testing - Python Script

Ran the attached case_test.py which uses SQLAlchemy inspection to get column information for existing tables. Target DBMS tested: Vector 6.3 database on Ubuntu and Ingres 11.2 database on Windows

case_test.py.txt

Example output

Python 3.10.7 (tags/v3.10.7:6cc6b13, Sep  5 2022, 14:08:36) [MSC v.1933 64 bit (AMD64)] on win32

2.0.28
ingres://testuser:TESTPWD@TESTMACHINE:21064/casedb
(['Driver={Actian II};Database=casedb;HostName=TESTMACHINE;ListenAddress=21064;UID=testuser;PWD=TESTPWD'], {})
('II 11.2.0 (a64.win/100)',)
('testuser                        ', 'otheruser                       ', '$ingres                         ')
('DB_DELIMITED_CASE               ', 'MIXED                           ')
('DB_NAME_CASE                    ', 'LOWER                           ')
('DB_REAL_USER_CASE               ', 'LOWER                           ')
('MIXEDCASE_NAMES                 ', 'N                               ')

============================================================================
================  Table names as stored with actual case  ==================
============================================================================
('TABLE UPPER DELIMITED ',)
('Table Mixed Delimited ',)
('table lower delimited ',)
('table_lower ',)
('table_mixed ',)
('table_upper ',)

=====================================================================================
=====  Table Names Inspections using SQLAlchemy-Ingres method: get_table_names  =====
=====================================================================================
['TABLE UPPER DELIMITED', 'table_lower', 'table lower delimited', 'table_upper', 'Table Mixed Delimited', 'table_mixed']

=======================================================
===============  Columns of each table  ===============
=======================================================
Columns for table ( TABLE UPPER DELIMITED ): [{'name': 'COL1 DELIM', 'nullable': True, 'default': None, 'type': Integer()}]
Columns for table ( table_lower ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}]
Columns for table ( table lower delimited ): [{'name': 'col1 delim', 'nullable': True, 'default': None, 'type': Integer()}]
Columns for table ( table_upper ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}]
Columns for table ( Table Mixed Delimited ): [{'name': 'Col1 Delim', 'nullable': True, 'default': None, 'type': Integer()}]
Columns for table ( table_mixed ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}]

Testing - Apache Superset

Python 3.10.12

apache-superset           3.1.1
pyodbc                    5.1.0
pypyodbc                  1.3.6
SQLAlchemy                1.4.51.dev0
sqlalchemy-ingres         0.0.7.dev1

Tested Superset's auto-retrieval of column metadata and row data in Superset SQL Lab, plus adhoc SQL statements in SQL Lab. Used backends of Ingres 11.2 on Windows and Vector 6.3 on Ubuntu. All sanity tests ran successfully, correctly retrieving the data.

[vsreeniv]

Hello Michael, Maybe you can remove the if / else condition and just return the variable "name", in both functions. Thanks & Best Regards Vibi

…

On Thu, Mar 21, 2024 at 8:52 PM Michael Habiger ***@***.***> wrote: Description Removed forced conversion of identifiers to lowercase so that uppercase and mixed-case scenarios now work properly. Also removed encoding/decoding of string to bytes. Proper handling (if needed) will be dealt with via #36 <#36>. Testing - Python Script Ran the attached case_test.py which uses SQLAlchemy inspection to get column information for existing tables. Target DBMS tested: Vector 6.3 database on Ubuntu and Ingres 11.2 database on Windows case_test.py.txt <https://github.com/ActianCorp/sqlalchemy-ingres/files/14706188/case_test.py.txt> Example output Python 3.10.7 (tags/v3.10.7:6cc6b13, Sep 5 2022, 14:08:36) [MSC v.1933 64 bit (AMD64)] on win32 2.0.28 ***@***.***:21064/casedb (['Driver={Actian II};Database=casedb;HostName=TESTMACHINE;ListenAddress=21064;UID=testuser;PWD=TESTPWD'], {}) ('II 11.2.0 (a64.win/100)',) ('testuser ', 'otheruser ', '$ingres ') ('DB_DELIMITED_CASE ', 'MIXED ') ('DB_NAME_CASE ', 'LOWER ') ('DB_REAL_USER_CASE ', 'LOWER ') ('MIXEDCASE_NAMES ', 'N ') ============================================================================ ================ Table names as stored with actual case ================== ============================================================================ ('TABLE UPPER DELIMITED ',) ('Table Mixed Delimited ',) ('table lower delimited ',) ('table_lower ',) ('table_mixed ',) ('table_upper ',) ===================================================================================== ===== Table Names Inspections using SQLAlchemy-Ingres method: get_table_names ===== ===================================================================================== ['TABLE UPPER DELIMITED', 'table_lower', 'table lower delimited', 'table_upper', 'Table Mixed Delimited', 'table_mixed'] ======================================================= =============== Columns of each table =============== ======================================================= Columns for table ( TABLE UPPER DELIMITED ): [{'name': 'COL1 DELIM', 'nullable': True, 'default': None, 'type': Integer()}] Columns for table ( table_lower ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}] Columns for table ( table lower delimited ): [{'name': 'col1 delim', 'nullable': True, 'default': None, 'type': Integer()}] Columns for table ( table_upper ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}] Columns for table ( Table Mixed Delimited ): [{'name': 'Col1 Delim', 'nullable': True, 'default': None, 'type': Integer()}] Columns for table ( table_mixed ): [{'name': 'col1', 'nullable': True, 'default': None, 'type': Integer()}] Testing - Apache Superset Python 3.10.12 apache-superset 3.1.1 pyodbc 5.1.0 pypyodbc 1.3.6 SQLAlchemy 1.4.51.dev0 sqlalchemy-ingres 0.0.7.dev1 Tested Superset's auto-retrieval of column metadata and row data in *Superset SQL Lab*, plus adhoc SQL statements in SQL Lab. Used backends of Ingres 11.2 on Windows and Vector 6.3 on Ubuntu. All sanity tests ran successfully, correctly retrieving the data. ------------------------------ You can view, comment on, or merge this pull request online at: #39 Commit Summary - 3d09cd7 <3d09cd7> fix identifier case handling #37 - b483e8c <b483e8c> Removed decode File Changes (1 file <https://github.com/ActianCorp/sqlalchemy-ingres/pull/39/files>) - *M* lib/sqlalchemy_ingres/base.py <https://github.com/ActianCorp/sqlalchemy-ingres/pull/39/files#diff-fd5cb38c20c2a27b0d1364dfd9908e23a30dc1ca7e9495da4c63dba924d425cc> (4) Patch Links: - https://github.com/ActianCorp/sqlalchemy-ingres/pull/39.patch - https://github.com/ActianCorp/sqlalchemy-ingres/pull/39.diff — Reply to this email directly, view it on GitHub <#39>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZ4C2CGYSBHEMPOGCYY3WHLYZM26FAVCNFSM6AAAAABFCDHEESVHI2DSMVQWIX3LMV43ASLTON2WKOZSGIYDCMBUGIYTMOA> . You are receiving this because your review was requested.Message ID: ***@***.***>