[DBAL-868] [GH-566] added support of LOB download Created: 16/Apr/14  Updated: 16/Apr/14

Status: Open
Project: Doctrine DBAL
Component/s: None
Affects Version/s: None
Fix Version/s: None
Security Level: All

Type: Bug Priority: Major
Reporter: Doctrine Bot Assignee: Benjamin Eberlei
Resolution: Unresolved Votes: 0
Labels: None


 Description   

This issue is created automatically through a Github pull request on behalf of bborrel:

Url: https://github.com/doctrine/dbal/pull/566

Message:

OCI8Statement::bindParam() was forcing a LOB upload when passing a
parameter of a LOB type. Also, this forced upload is out of concern for
OCI8Statement::bindParam().

So I removed it to support LOB download. Now call of
oci_new_descriptor() to create a LOB is the responsibility of the
controller. To help this call I added OCI8Connection::getDBH() to be
able to get the resource to the connection.

File download example:
```
$connection = new \Doctrine\DBAL\Connection(...);
$sql = 'BEGIN APP.PKG_NAME.GET_FILE(:id,:data); END;';
$stmt = $connection->prepare($sql);
$stmt->bindValue('id', $id);

$blob = oci_new_descriptor($connection->getWrappedConnection()->getDBH(), OCI_DTYPE_LOB);
$stmt->bindParam('data', $blob, PDO::PARAM_LOB);

$stmt->execute();
$data = $blob->load();
```

File upload example:
```
$connection = new \Doctrine\DBAL\Connection(...);
$sql = 'BEGIN APP.PKG_NAME.PUT_FILE(:id,:bData); END;';
$stmt = $connection->prepare($sql);
$stmt->bindValue('id', $id);

$blob = oci_new_descriptor($connection->getWrappedConnection()->getDBH(), OCI_DTYPE_LOB);
$blob->writeTemporary($data, OCI_TEMP_BLOB);
$stmt->bindParam('data', $blob, PDO::PARAM_LOB);

$stmt->execute();
```






[DBAL-867] Doctrine\DBAL\Driver\Connection should be marked as an internal interface Created: 16/Apr/14  Updated: 16/Apr/14

Status: Open
Project: Doctrine DBAL
Component/s: None
Affects Version/s: None
Fix Version/s: None
Security Level: All

Type: Documentation Priority: Major
Reporter: Christophe Coevoet Assignee: Benjamin Eberlei
Resolution: Unresolved Votes: 0
Labels: None


 Description   

Currently, the main entry point Doctrine\DBAL\Connection is documented as a wrapper around Doctrine\DBAL\Driver\Connection.
This is very misleading and encourages other library to typehint against Doctrine\DBAL\Driver\Connection rather than Doctrine\DBAL\Connection. See https://github.com/symfony/symfony/pull/10720 for the original discussion.

However, the discussion in https://github.com/doctrine/dbal/pull/414#discussion_r7688765 implies that they should actually not be related together (but it cannot be fixed for BC reasons). The phpdoc should at least be changed






[DBAL-866] Foreign Key Constraints does not work with Doctrine/Symfony and SQLite Created: 12/Apr/14  Updated: 12/Apr/14

Status: Open
Project: Doctrine DBAL
Component/s: None
Affects Version/s: 2.4.2
Fix Version/s: None
Security Level: All

Type: Bug Priority: Major
Reporter: Christian Stoller Assignee: Benjamin Eberlei
Resolution: Unresolved Votes: 0
Labels: None
Environment:

PHP 5.5.9, SQLite3 module version 0.7-dev, SQLite Library 3.8.3.1



 Description   

I have posted a question on stackoverflow already to get help on this issue, but nobody could give me a sufficient answer. See here.

#370 says that support for foreign keys support for SQLite has been implemented. But in my case it does not work. I have defined two entities:

Unable to find source-code formatter for language: yaml. Available languages are: actionscript, html, java, javascript, none, sql, xhtml, xml
Category:
    type: entity
    id:
        id:
            type: integer
            id: true
            generator:
                strategy: AUTO
    oneToMany:
        ministries:
            targetEntity: Ministry
            cascade: [persist]
            mappedBy: category

And

Unable to find source-code formatter for language: yaml. Available languages are: actionscript, html, java, javascript, none, sql, xhtml, xml
Ministry:
    type: entity
    id:
        id:
            type: integer
            id: true
            generator:
                strategy: AUTO
    manyToOne:
        category:
            targetEntity: Category
            inversedBy: ministries
            joinColumn:
                name: category_id
                nullable: false
                onDelete: CASCADE

But when I delete a category, the ministry entities do not get deleted, although the constraint should cascade. What am I missing?

Do I have to configure anything to get that working or is it a bug?






[DBAL-865] [GH-565] fix lastInsertId typehint in SqlSrv Created: 12/Apr/14  Updated: 13/Apr/14  Resolved: 13/Apr/14

Status: Resolved
Project: Doctrine DBAL
Component/s: None
Affects Version/s: None
Fix Version/s: 2.5
Security Level: All

Type: Bug Priority: Major
Reporter: Doctrine Bot Assignee: Marco Pivetta
Resolution: Fixed Votes: 0
Labels: None


 Description   

This issue is created automatically through a Github pull request on behalf of Tobion:

Url: https://github.com/doctrine/dbal/pull/565

Message:

An LastInsertId|null is expected.



 Comments   
Comment by Doctrine Bot [ 13/Apr/14 ]

A related Github Pull-Request [GH-565] was closed:
https://github.com/doctrine/dbal/pull/565

Comment by Marco Pivetta [ 13/Apr/14 ]

Merged: https://github.com/doctrine/dbal/commit/32817aac490eeba4c39bd7670ec2f6182e496fc5





[DBAL-864] Failure to insert FALSE into a bool column Created: 10/Apr/14  Updated: 11/Apr/14

Status: Open
Project: Doctrine DBAL
Component/s: None
Affects Version/s: None
Fix Version/s: None
Security Level: All

Type: Bug Priority: Major
Reporter: Matthieu Napoli Assignee: Benjamin Eberlei
Resolution: Unresolved Votes: 0
Labels: None


 Description   

I have experienced this problem with MySQL, I am not sure how it behaves with other platforms. Also, maybe this duplicates http://www.doctrine-project.org/jira/browse/DBAL-630 but since that issue is specifically about PostreSQL I am creating a separate one.

[Doctrine\DBAL\Exception\DriverException]
An exception occurred while executing 'INSERT INTO ACL_Authorization
(role_id, securityIdentity_id, parentAuthorization_id, entity_class, entity_id, cascadable)
VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)'
with params [2, 2, null, "Account\\Domain\\Account", 2, false]:

SQLSTATE[HY000]: General error: 1366 Incorrect integer value: '' for column 'cascadable' at row 1

I think it's related to https://bugs.php.net/bug.php?id=49255 (PDO fails to insert boolean FALSE to MySQL in prepared statement) which casts FALSE into an empty string.



 Comments   
Comment by Marco Pivetta [ 10/Apr/14 ]

Matthieu Napoli this issue could need some additional environment information. Also, isn't there a parameter count mismatch?

Comment by Steve Müller [ 10/Apr/14 ]

Yeah the VALUES clause looks weird concerning number of parameters. Could you maybe also provide information of how you constructed the query? DBAL? ORM? A little code snipüpet would help...

Comment by Matthieu Napoli [ 10/Apr/14 ]

I removed useless values to clear up the message, don't mind the excessive "?" in "VALUES".

Here is the code that trigger this: https://github.com/myclabs/ACL/blob/master/src/Repository/AuthorizationRepository.php#L62

More explicitly, this is: $connection->insert($tableName, $data) with $data being a simple array.

We are talking about DBAL (else I would have opened the issue in the ORM project), probably master (my constraint is master of ORM).

Regarding the environment, this is weird: I can't reproduce it on Ubuntu (PHP 5.5, MySQL version I don't know). The bug appears on OS X, PHP 5.5.5, MySQL 5.6.17 (just upgraded).

Comment by Matthieu Napoli [ 10/Apr/14 ]

I confirm that this is related to FALSE being casted to string, when I cast the boolean to an int it works. Example:

$data = [
    // ...
    'cascadable' => (int) $authorization->isCascadable(),
];
Comment by Matthieu Napoli [ 10/Apr/14 ]

And to be extra-sure I tried casting to boolean, but I still get the error:

$data = [
    // ...
    'cascadable' => (bool) $authorization->isCascadable(),
];
Comment by Marco Pivetta [ 11/Apr/14 ]

Matthieu Napoli is this due to a change in the ORM, an upgrade on your side or are were you implementing something in your codebase? I just wanted to be sure if this may be due to a breakage on your side or something you're experiencing on your code changes.

Comment by Matthieu Napoli [ 11/Apr/14 ]

There is nothing "new" on my side except the code (I mean I didn't "upgrade" anything): this is a new project I started.

Since I use embedded objects, I required doctrine/orm dev-master (or 2.5-BETA3 I don't know but it's roughly the same). Then I used DBAL to do a simple insert:

$data = [
    ...
    'cascadable' => $authorization->isCascadable(),
];

$connection->insert($tableName, $data);

The tests for this "ACL" project are run using SQLite in memory, and they always pass (on every machine).

When I use the project (as a dependency) in another one, with a MySQL backend, it works (i.e. no error) on my Ubuntu machine but not on my OS X machine.

I will be trying to reproduce it in a test today, however I am on Ubuntu right now (work) so maybe I won't see it.

(side note: I have no idea what's the deal between the Ubuntu and OS X machine, both have PHP 5.5 and a latest version of MySQL...)

Comment by Matthieu Napoli [ 11/Apr/14 ]

So as I feared, the test I wrote passed on my Ubuntu machine but fails on my Macbook.

Here is the test: https://github.com/mnapoli/dbal/compare/DBAL-864 As you can see it's as simple as it can be.

Exception : [Doctrine\DBAL\Exception\DriverException] An exception occurred while executing 'INSERT INTO dbal864tbl (foo) VALUES (?)' with params [false]:

SQLSTATE[HY000]: General error: 1366 Incorrect integer value: '' for column 'foo' at row 1

With queries:
3. SQL: 'INSERT INTO dbal864tbl (foo) VALUES (?)' Params: ''
2. SQL: 'CREATE TABLE dbal864tbl (foo TINYINT(1) NOT NULL) DEFAULT CHARACTER SET utf8 COLLATE utf8_unicode_ci ENGINE = InnoDB' Params:

The test passes with SQLite however…

Comment by Matthieu Napoli [ 11/Apr/14 ]

I am confused by the explanation given in https://bugs.php.net/bug.php?id=49255 but I tend to think it's related. When I run the test in debug and step by step, I confirm that the data passed to PDO is array(false). The casting of false to '' happens inside PDO.

Edit: It's starting to make sense, have a look here: https://bugs.php.net/bug.php?id=33876

The PDO documentation says that PDOStatement::execute says that "All values are treated as PDO::PARAM_STR" (http://php.net/manual/en/pdostatement.execute.php), whereas this should work:

$res->bindValue(1, false, PDO_PARAM_BOOL);
$res->execute();




Generated at Thu Apr 17 03:59:34 UTC 2014 using JIRA 6.2.3#6260-sha1:63ef1d6dac3f4f4d7db4c1effd405ba38ccdc558.