Support Informix #436

MMatten · 2015-10-25T19:39:22Z

Support for Informix (using alternate approach of allowing StatementExecution sub-classing instead of adding more parameters to existing class methods and more property getters on the DBEnvironment interface).

TODO:

Simplify testing expected exceptions
Test the class of the Big Decimal transform() output. (The reason which forced us to have custom handling for Informix).
Decide about the big decimal to exact java.math.BigDecimal transformer name
Some refactoring - db accessor set method
DbParameterAccessor: select the name of the transform to JDBC - compatible type method
Rename JDBC to Jdbc for the newly created fields and classes

… still not working) -moved setAutoCommit to AbstractDbEnvironment.afterConnectionEstablished -added calls to getAutoCommit before calling setAutoCommit -removed overrides to commit / afterConnectionEstablished in InformixEnvironment

…e made

javornikolov · 2015-12-01T21:44:38Z

dbfit-java/oracle/src/main/java/dbfit/environment/OracleEnvironment.java

                paramDirection, getSqlType(dataType), getJavaClass(dataType),
                paramPosition, normaliseTypeName(dataType), userTypeName);

        return dbp;
    }

+    private OracleDbParameterAccessor createOracleDbParameterAcccessor(


Would it work if we just override createDbParameterAcccessor here? (Just to be keep things like in other envs).

We can overload createDbParameterAcccessor (the Oracle env method has the extra String userTypeName param) just to give this the same name as in AbstractDbEnvironment.

OK, let's rename and add @Override annotation.

I can rename/overload but not add the annotation as we're not overriding the super-type's method.

Ah, we have different list of parameters. Then I think it would be better to keep it as it is - with Oracle in the name (just to make it more obvious that it's a bit different).

javornikolov · 2015-12-01T21:51:00Z

dbfit-java/postgres/src/main/java/dbfit/environment/PostgresEnvironment.java

@@ -245,7 +247,8 @@ public Class getJavaClass(String dataType) {
            if (!dataType.equals("void")) {
                allParams.put("", new DbParameterAccessor("",


Also here - createDbParameterAccessor vs new ...

…hod names

javornikolov · 2015-12-01T21:59:57Z

dbfit-java/core/src/test/java/dbfit/util/DbParameterAccessorsMapBuilderTest.java

@@ -1,5 +1,6 @@
 package dbfit.util;

+import dbfit.util.TypeTransformer;


I guess this import is also unused.

MMatten · 2015-12-01T22:08:22Z

Ok. All applied.

MMatten · 2015-12-01T22:09:04Z

Ah, need to revert that overloading!

…lign method names

javornikolov · 2015-12-01T22:18:02Z

dbfit-java/core/src/main/java/dbfit/util/DbParameterAccessorsMapBuilder.java

@@ -14,9 +14,9 @@
        return parameterAccessors;
    }

-    public void add(String name, Direction direction, int sqlType, Class javaType) {
+    public void add(String name, Direction direction, int sqlType, Class javaType, TypeTransformerFactory dbfitToJdbcTransformerFactory) {


This TypeTransformerFactory dbfitToJdbcTransformerFactory may have been moved as attribute (and initialized via the constructor) instead of passing it as parameter.
But no urgency to tweak that. Can be revised some other time.

I'm just about to push a commit to change this...

javornikolov · 2015-12-01T22:25:37Z

I think we're already in relatively good shape here. As we proceed towards merging the stuff:

Maybe it makes sense to push some of the refactoring (the support of adding toJdbcCompatible transformers) as separate PR? It may be helpful to review it later for further improvements (or if some regression emerges /though I hope we don't have any/).
Also consider squashing the commits.

…transformer via contructor

MMatten · 2015-12-02T16:03:36Z

Maybe it makes sense to push some of the refactoring (the support of adding toJdbcCompatible transformers) as separate PR?

Shall we try to slice this off and merge it before the main Informix changes?

MMatten · 2015-12-02T16:26:03Z

Shall we try to slice this off and merge it before the main Informix changes?

If "yes", then can you give me a few pointers on how to do that?

MMatten · 2015-12-02T16:38:35Z

Well, let me have a crack at this

http://stackoverflow.com/questions/1628563/move-the-most-recent-commits-to-a-new-branch-with-git

...first.

javornikolov · 2015-12-02T18:12:07Z

Shall we try to slice this off and merge it before the main Informix changes?
If "yes", then can you give me a few pointers on how to do that?

Yes. If we're about to squash the commits - then one approach is:

create a new branch from master
git merge --squash --no-commit
Then remove the files not related to what we're moving there (git checkout -- path). And then commit.

If history is about to be preserved: some variation of git rebase can be used to select which commits we want to move.

MMatten · 2015-12-02T21:04:19Z

We've touched AbstractDbEnvironment and DBEnvironment for both the 'JDBC compatible value' change and the change to create the StatementExecution sub-types (required for the Informix stuff).

I guess I'll have to go looking for the commit where this direction (the 2nd change) was started. Or do you know another way of approaching this?

MMatten · 2015-12-02T21:12:31Z

Sorry, make that just AbstractDbEnvironment.

MMatten · 2015-12-02T21:24:54Z

The docs for the --no-squash variant say it commits so no chance of even using rebase -i.

MMatten · 2015-12-02T21:37:22Z

I guess I can just undo that change on that one file and deal with the conflict later when merging the Informix stuff.

MMatten · 2015-12-02T21:54:03Z

...and DbParameterAccessor.

javornikolov · 2015-12-02T21:56:39Z

We've touched AbstractDbEnvironment and DBEnvironment for both the 'JDBC compatible valuechange and the change to create theStatementExecution` sub-types (required for the Informix stuff).

I guess I'll have to go looking the commit where this direction was started. Or do you know another way of approaching this?

Right... I've forgotten about that create statement execution. Makes sense to go with separate PR for it too. So just merge --squash would be too bulky perhaps. Interactive rebase might be the easiest way to go.

MMatten · 2015-12-02T22:13:07Z

Part one created on #448.

MMatten and others added 30 commits October 3, 2015 21:33

Add disk space to test VM

fb8316a

Install Informix IDS if package available

7d40e81

Hide output when detecting existing dbfit database

acb51af

Fix deinstaller, add docs

988a487

Fixes to Informix installer/deinstaller

e795a3e

adding Informix wrapper

6ede1e4

update gradle project to add Informix JDBC jar compile-time reference

7c90e39

removed informix stuff so a pull request with only core changes can b…

75562f9

…e made

remove informix from gradle settings.gradle

477357b

removed some more changes not needed in fork

60243dd

putting back the Informix changes

0307416

added handling for other datatypes

ba9ee11

latest changes to help with named parameters in informix

129b285

Apply Java code style - whitespace tidy

30994ed

Exit Informix installer gracefully when no installer available

6573b40

More fixes and tidies

df80cff

Get core tests passing

43f8a10

Corrections to doc and build

62d8b6f

Apply project code style to Java code

13e9ec6

Refine Informix JDBC driver file names

7167873

Fix whitespace to apply standard Java style

0989a86

Fix whitespace to apply standard Java style

f8d8c1b

Add instruction for manually starting/stopping the Informix server

2ece601

Use connection properties file for StandaloneFixtures tests

bc387c0

Remove debugging output

3e23359

Use connection properties file for StandaloneFixtures tests

7ba4897

Revert VM to 1100MB RAM after more testing of Ora+DB2+Informix

8678c79

Allow Cheffile.lock to update

76b6226

Run tests in ANSI and Informix modes

dc41e63

javornikolov reviewed Dec 1, 2015
View reviewed changes

Create DbParameterAccessor object from createDbParameterAccessor

bff0e18

javornikolov reviewed Dec 1, 2015
View reviewed changes

Overload AbstractDbEnvironment.createDbParameterAccessor to align met…

83eb138

…hod names

javornikolov reviewed Dec 1, 2015
View reviewed changes

MMatten added 3 commits December 1, 2015 22:02

Create DbParameterAccessor object from createDbParameterAccessor

bfa94c5

Remove unused imports

b904ab2

Remove unused imports

8afa715

Revert: Overload AbstractDbEnvironment.createDbParameterAccessor to a…

43c9081

…lign method names

javornikolov reviewed Dec 1, 2015
View reviewed changes

Initialise DbParameterAccessorsMapBuilder object's dbfitToJdbc value …

54dcdf6

…transformer via contructor

This was referenced Dec 28, 2015

Allow customisation of RDBMS function/procedure execution #451

Merged

Support IBM Informix #452

Merged

javornikolov closed this in #452 Dec 31, 2015

javornikolov added the enhancement label Jan 28, 2018

javornikolov added this to the 4.0.0 milestone Jan 28, 2018

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support Informix #436

Support Informix #436

MMatten commented Oct 25, 2015

javornikolov Dec 1, 2015

MMatten Dec 1, 2015

javornikolov Dec 1, 2015

MMatten Dec 1, 2015

javornikolov Dec 1, 2015

javornikolov Dec 1, 2015

javornikolov Dec 1, 2015

MMatten commented Dec 1, 2015

MMatten commented Dec 1, 2015

javornikolov Dec 1, 2015

MMatten Dec 2, 2015

javornikolov commented Dec 1, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

javornikolov commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

javornikolov commented Dec 2, 2015

MMatten commented Dec 2, 2015

		@@ -245,7 +247,8 @@ public Class getJavaClass(String dataType) {
		if (!dataType.equals("void")) {
		allParams.put("", new DbParameterAccessor("",

		@@ -1,5 +1,6 @@
		package dbfit.util;

		import dbfit.util.TypeTransformer;

Support Informix #436

Support Informix #436

Conversation

MMatten commented Oct 25, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MMatten commented Dec 1, 2015

MMatten commented Dec 1, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javornikolov commented Dec 1, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

javornikolov commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

MMatten commented Dec 2, 2015

javornikolov commented Dec 2, 2015

MMatten commented Dec 2, 2015