DPY-3040 on AR8MSWIN1256 (charset id 560) when reading DbObject string attributes in Thin mode

Aberta
  • Algo errado?
#636 1 comentário 0 reações 0 responsáveis Ver no GitHub

Ninguém assumiu esta issue ainda.

Avaliação

Dificuldade
3/5
Tempo estimado
1-2 dias
Facilidade para iniciantes
57/100
Tipo de issue
Bug
Clareza
Razoavelmente clara
Status de atividade
Ativa
Stack de tecnologia
python
Domínio
backend, database

Direção de pesquisa

Start with src/oracledb/impl/thin/capabilities.pyx, where the issue says ORACLE_CHARSET_TO_PYTHON_ENCODING maps cp1256 to charset id 175, while the database reports id 560. Check the neighboring Windows code-page mappings (172–177) and reproduce with the provided DbObject OUT-bind example on a database using AR8MSWIN1256. Done when reading the string attribute succeeds in Thin mode and the neighboring mappings are checked for the reported mismatches.

Escrita pelo modelo de indexação a partir do texto da issue.

Descrição

bug
  1. What versions are you using?

Oracle Database: 19.31.0.0.0
Database character set: AR8MSWIN1256 (SELECT NLS_CHARSET_ID('AR8MSWIN1256') FROM dual returns 560)

platform.platform: Windows-2022Server-10.0.20348-SP0
sys.maxsize > 2**32: True
platform.python_version: 3.12.13
oracledb.__version__: 4.0.2
  1. Is it an error or a hang or a crash?

An error.

  1. What error(s) or behavior you are seeing?

Ordinary query columns work fine on this database, because the server converts them. Reading a VARCHAR2 attribute of a DbObject (here an element of a nested-table OUT bind) fails:

  File "...\oracledb\dbobject.py", line 46, in __getattr__
    return self._impl.get_attr_value(attr_impl)
  File "src/oracledb/impl/thin/dbobject.pyx", line 483, in oracledb.thin_impl.ThinDbObjectImpl.get_attr_value
  File "src/oracledb/impl/thin/dbobject.pyx", line 198, in oracledb.thin_impl.ThinDbObjectImpl._ensure_unpacked
  File "src/oracledb/impl/thin/dbobject.pyx", line 316, in oracledb.thin_impl.ThinDbObjectImpl._unpack_data
  File "src/oracledb/impl/thin/dbobject.pyx", line 349, in oracledb.thin_impl.ThinDbObjectImpl._unpack_data_from_buf
  File "src/oracledb/impl/thin/dbobject.pyx", line 406, in oracledb.thin_impl.ThinDbObjectImpl._unpack_value
  File "src/oracledb/impl/thin/capabilities.pyx", line 164, in oracledb.thin_impl.Capabilities._get_encoding
  File "...\oracledb\errors.py", line 199, in _raise_err
    raise error.exc_type(error) from cause
oracledb.exceptions.NotSupportedError: DPY-3040: database character set id 560 is not supported by python-oracledb in thin mode

In src/oracledb/impl/thin/capabilities.pyx, ORACLE_CHARSET_TO_PYTHON_ENCODING maps cp1256 (AR8MSWIN1256) to id 175, but the database reports AR8MSWIN1256 as id 560. So the lookup misses even though the encoding is meant to be supported. The neighbouring Windows code-page entries (172–177) may also be worth checking. On our database:

SELECT cs, NLS_CHARSET_ID(cs) AS id
  FROM (SELECT COLUMN_VALUE cs
          FROM TABLE(sys.odcivarchar2list('EL8MSWIN1253', 'TR8MSWIN1254', 'IW8MSWIN1255',
                                          'AR8MSWIN1256', 'BLT8MSWIN1257', 'VN8MSWIN1258')));

"CS"	"ID"
"EL8MSWIN1253"	174
"TR8MSWIN1254"	177
"IW8MSWIN1255"	175
"AR8MSWIN1256"	560
"BLT8MSWIN1257"	179
"VN8MSWIN1258"	45

Workaround: return the collection as a REF CURSOR over TABLE(...) instead of binding it as an object, so the server does the conversion.

  1. Does your application call init_oracle_client()?

No, it uses Thin mode.

  1. Include a runnable Python script that shows the problem.

Needs a database whose character set is AR8MSWIN1256.

CREATE OR REPLACE TYPE dpy3040_t AS OBJECT (id NUMBER, name VARCHAR2(50));
/
CREATE OR REPLACE TYPE dpy3040_tab AS TABLE OF dpy3040_t;
/

import oracledb

conn = oracledb.connect(user="<user>", password="<password>", dsn="<dsn>")
print("db version:", conn.version)

tab_type = conn.gettype("DPY3040_TAB")
with conn.cursor() as cur:
    out = cur.var(tab_type)
    cur.execute("""
        BEGIN
            :out := dpy3040_tab(dpy3040_t(1, 'abc'));
        END;""", out=out)
    element = out.getvalue().aslist()[0]
    print(element.ID)    # works: NUMBER needs no character set
    print(element.NAME)  # DPY-3040: database character set id 560 is not supported
Linguagem predominante
Python
Estrelas
458
Forks
123
Nenhum PR com merge em 30d

Preparar o ambiente

Primeiros passos

  1. Leia a issue inteira e depois o guia de contribuição do projeto.
  2. Comente na issue dizendo que vai assumir — evita que duas pessoas façam o mesmo trabalho.
  3. Faça um fork do repositório e trabalhe em uma branch.
  4. Abra um pull request que referencie o número da issue.

Mais de oracle/python-oracledb

Todas as issues de oracle/python-oracledb

Issues semelhantes

Mais issues de Python

Receba novas issues na sua caixa de entrada

Um resumo curto de issues do GitHub para quem está começando.