Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add support for UTF-8 feature extension. #722

Merged
merged 14 commits into from
Jun 22, 2018
Merged

Conversation

ulvii
Copy link
Contributor

@ulvii ulvii commented Jun 13, 2018

Support for UTF-8 feature extension that will be introduced in newer versions of SQL Server.

@@ -4419,6 +4441,8 @@ else if (serverMajorVersion >= 9) // Yukon (9.0) --> TDS 7.2 // Prelogin disconn

len2 = len2 + 1; // add 1 to length becaue of FeatureEx terminator

len2 = len2 + writeUTF8SupportFeatureRequest(false, tdsWriter);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a little convoluted, the function basically always returns "5". It also has a Boolean to turn on/off the actual writing. Shouldn't we remove the length calculation from this function and just have a static variable which equals 5? Can also remove the Boolean parameter.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is done to keep the same structure for all feature extension requests, in case we calculate the length differently in the future. Take a look at writeFedAuthFeatureRequest, writeAEFeatureRequest, writeDataClassificationFeatureRequest, The methods are implemented the same way in .NET driver too.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It makes sense with fedAuth because the length is actually variable. For the others, it seems there's a set value(6 for AE, 5 for UTF8, 6 for data classification). I guess it's fine if we're consistent, but I'm just saying that other than the fedAuth function, the others don't really make a lot of sense.

@@ -4598,6 +4622,8 @@ else if (serverMajorVersion >= 9) // Yukon (9.0) --> TDS 7.2 // Prelogin disconn
writeFedAuthFeatureRequest(true, tdsWriter, fedAuthFeatureExtensionData);
}

writeUTF8SupportFeatureRequest(true, tdsWriter);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Length is being "calculated" and returned, but we don't use the value. Seems like we're always using half of the functionalities.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes I've never liked this. It's not really "calculated" at all the whole thing is just hardcoded

public static void setUp() throws ClassNotFoundException, SQLException {
databaseName = RandomUtil.getIdentifier("UTF8Database");
tableName = AbstractSQLGenerator.escapeIdentifier(RandomUtil.getIdentifier("RequestBoundaryTable"));
connection = PrepUtil.getConnection(getConfiguredProperty("mssql_jdbc_test_connection_properties"));
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

who closes this connection?

@codecov-io
Copy link

codecov-io commented Jun 13, 2018

Codecov Report

Merging #722 into dev will decrease coverage by 0.08%.
The diff coverage is 50%.

Impacted file tree graph

@@             Coverage Diff              @@
##                dev     #722      +/-   ##
============================================
- Coverage     48.12%   48.03%   -0.09%     
+ Complexity     2629     2626       -3     
============================================
  Files           113      113              
  Lines         26631    26648      +17     
  Branches       4472     4477       +5     
============================================
- Hits          12816    12801      -15     
- Misses        11695    11721      +26     
- Partials       2120     2126       +6
Flag Coverage Δ Complexity Δ
#JDBC42 47.52% <50%> (-0.11%) 2576 <2> (-4)
#JDBC43 47.98% <50%> (-0.01%) 2624 <2> (+6)
Impacted Files Coverage Δ Complexity Δ
...om/microsoft/sqlserver/jdbc/SQLServerResource.java 100% <ø> (ø) 4 <0> (ø) ⬇️
...in/java/com/microsoft/sqlserver/jdbc/IOBuffer.java 52.68% <0%> (-0.06%) 0 <0> (ø)
...ava/com/microsoft/sqlserver/jdbc/SQLCollation.java 92.1% <50%> (-0.43%) 14 <0> (ø)
.../microsoft/sqlserver/jdbc/SQLServerConnection.java 48.01% <53.84%> (+0.02%) 323 <2> (+2) ⬆️
...rc/main/java/com/microsoft/sqlserver/jdbc/DDC.java 43.84% <0%> (-1.57%) 107% <0%> (-1%)
...c/main/java/com/microsoft/sqlserver/jdbc/Util.java 60.91% <0%> (-0.44%) 88% <0%> (-1%)
...m/microsoft/sqlserver/jdbc/SQLServerResultSet.java 33.32% <0%> (-0.4%) 247% <0%> (-4%)
...n/java/com/microsoft/sqlserver/jdbc/Parameter.java 62.47% <0%> (-0.21%) 63% <0%> (ø)
...om/microsoft/sqlserver/jdbc/SQLServerBulkCopy.java 52.61% <0%> (-0.13%) 241% <0%> (ø)
...rc/main/java/com/microsoft/sqlserver/jdbc/dtv.java 63.24% <0%> (-0.12%) 0% <0%> (ø)
... and 2 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c34e1a9...72d3b9b. Read the comment docs.

rene-ye
rene-ye previously approved these changes Jun 13, 2018

if (1 > data.length) {
if (connectionlogger.isLoggable(Level.SEVERE)) {
connectionlogger.severe(toString() + " Unknown value for UTF8 support.");
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

use R_unknownUTF8SupportValue here too?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

no, we don't localize the logger messages.

@ulvii ulvii added this to the 6.5.4 milestone Jun 15, 2018
rene-ye
rene-ye previously approved these changes Jun 15, 2018
lilgreenbird
lilgreenbird previously approved these changes Jun 15, 2018
cheenamalhotra
cheenamalhotra previously approved these changes Jun 21, 2018
peterbae
peterbae previously approved these changes Jun 21, 2018
@cheenamalhotra cheenamalhotra dismissed stale reviews from peterbae, lilgreenbird, rene-ye, and themself via 72d3b9b June 21, 2018 19:45
@ulvii ulvii merged commit abc9851 into microsoft:dev Jun 22, 2018
@ulvii ulvii deleted the utf8Support branch June 22, 2018 22:11
cheenamalhotra added a commit that referenced this pull request Jun 30, 2018
* 623 fix

* 623 change stash

* Prepared Statement Caching fix for 'handle not found' errors

* Fix for PS Caching issue - Calling reset instead of on type def changes

* Updated comparison

* Change back assert check.

* Adding call to removeReference back + Fix for Batch processes intermittent failures.

* Removed DBName and made changes to resetPrepStmtHandle method

* Check for null handle before proceed

* Adding Old Constrcutor back to AKV Implementation

* Making baseURL final

* Remove unnecessary code.

* Use Bulk Copy API for batch insert operation

* Parse bug fixing and test added

* bug fix + additional tests

* change reflection for testing

* more test changes

* Add parsing logic for -- comment

* refactoring

* Update snapshot

* Bug fix / testing change

* Reflect comment change

* Feature | AKV Old Constructor changes - Reformatted code + Deprecated old Constructor and added a new constructor with 1 param

* Mark computed columns as IS_GENERATEDCOLUMN in the result set returned by getColumns() (#695)

* Fix | getColumns() API, changed column name from SS_IS_COMPUTED to IS_AUTOINCREMENT per JDBC specs | issue #600

* Fix | getColumns() API, changed column name from SS_IS_COMPUTED to IS_GENERATEDCOLUMN per JDBC specs | issue #600

* fix issue with redirection

* Fix | PS Caching - Remove commented lines

* Trigger Appveyor test

* Fix | AKV Old Constructor - Calling the other constructor instead.

* Fix | Reversed null checks

* Resolving alignment problems and comments

* Refactor two Bulk files into a common parent

* javadoc changes

* Applied formatter

* fix problem with precision / scale

* Fix | Fix some of the Javadoc warnings (#702)

* fix issue with setting all to true

* Resolved maven build warnings and java warnings regarding deprecated API (#701)

* Resolving maven warnings

* Removing jreVersion property

Does not make sense now that we use final name in the build itself. Only used in 1 place, hard-coding java version for different builds as that's what it represents anyways.

* java warnings

* make bamoo fixes

* resource bundle for junit test error strings (#698)

resource bundle for error message strings in junit tests

* undo some changes made to SQLServerConnection

* apply resource bundling changes

* Add support for JDK 10 in both Maven and Gradle (#691)

* Feature | Added support for JDK 10 in both Maven and Gradle - builds jre10 jars for the driver, replacing jre9

* JDK 10 | Merge 42 classes to base classes to reduce class redundancy.

* JDK 10 | Attempt to run JDK 10 with Appveyor

* Remove unwanted space

* Updating Travis script to use JDK 10

* Testing without addons

* Update script for Jacoco report to build 43 profile

* Revert driver changes for 42 compliance - to be added in a separate PR

* Revert Test class changes for 42 compliance - to be done in a separate PR

* Reformatted code

* Add ID to jacoco plugin execution task

* Kerberos Constrained Delegation Impersonated Credential Expiry fix (#636)

fix for automatic credential discarding

* update felix to 3.5.0

* Revised implementation

Decided to not dispose user created credentials at all.

* Updated flag set location

* changes for 6.5.3 preview release

* Revert "changes for 6.5.3 preview release"

This reverts commit 5c6ccd3.

* Changes in preparation for 6.5.3 preview release (#710)

* changes for preview release

* requested changes

* jre version update changes

* snapshot updates post release

* remove on_dw, and remove redundant fmtonly

* formatting

* fix for getSchema when using "-" in name

* Reformatting + adding more tests

* inherit the connection property in statement + fix issue with null / empty string being passed in as values

* Request Boundary methods - beginRequest()/endRequest() implementation (#708)

* Add | Request Boundary Methods - beginRequest()/endRequest() implementation

* Fix | Remove unused import from AbstractTest

* Fix | Applying review comments

* Fix | Moving RequestBoundaryMethodsTest.java to connection package

* added error message in resource file and changed files accordingly

* comment revisions

* use TestResource

* test changes

removed finals
removed database creation tracking

* drop database before creating

* replaced dropDBIfExists with Utils function

* added try-with-resources nest

avoid manually closing statements, and safetly handles resources.

* Fixing logic / adding more tests

* dont use test database in tests

* Change exception handling as per JDBC specs

* Add | Add missing license headers (#725)

* remove some comments

* Enable verify data (#724)

Fix to enable data verification in Junit tests. Also addresses intermittent failures with Time/Timestamp where the precision was being inaccurately judged.

* Fix | Refactored socket creation to simplify handling of socket creation

Refactors socket creation in SocketFinder.findSocket(...) to simplify handling of socket creation.

When the host resolves to a single address the driver now defers to getConnectedSocket(...)
to create the socket without spawning any threads. This happens regardless of whether we're
running on an IBM JDK. Previously the single address case would still use NIO on an IBM JDK.

On non-IBM JDKs the driver now handles both IPv4 and IPv6 addresses concurrently with a single
shared timeout. Previously hosts that resolved to both types of addresses were allowed half the
timeout for socket creation per address type with the resolution performed sequentially.

* reflect comments

* Add support for UTF-8 feature extension. (#722)

* Add | Support for UTF8 changes

* changed how logger works, refactored code in SQLServerBulkCommon due to that, changed exception being thrown to BatchUpdateException, added same logic for parsing in executeLargeBatch, and added tests accordingly.

* add more tests, make the prepared statement property go away

* Feature | Introduce support for "Data Classification Specifications" on fetched resultsets (#709)

* Feature | Data Classification Project | Phase 1 (contains temporary skipping 2 bytes)

* Feature | Data Classification - Removing extra bytes added before

* Feature | Data Classification - Added new test class for testing Data Classification support in the driver

* Remove one println

* Feature | Repackaged newly added files for Data Classification + improvements in source code

* Feature | Changing tokens to bytes instead of int

* Feature | Making variables private

* Formatted code + dropTable method called from Utils

* Feature | Data Classification - Changes as per review comments

* Fix | Review comment changes

* Change exception codes to follow series

* Fix Conflict issue

* Added missing Javadocs and headers for all new files

* Added getter/setter public for the useBulkCopyForBatchInsert connection property.

* Change implementation of child classes a bit

* Remove dependencies from tests that are from outside required libraries

* also remove hex from DBTable

* Fix bamboo problem + refactor test code

* Replace all connection and statements with try blocks

* change spacing

* refactor code

* refactoring

* Fix | Making driver default compliant to JDBC 4.2 Specs and update ADAL4J dependency to 1.6.0 (#711)

* Feature | Added support for JDK 10 in both Maven and Gradle - builds jre10 jars for the driver, replacing jre9

* JDK 10 | Merge 42 classes to base classes to reduce class redundancy.

* JDK 10 | Attempt to run JDK 10 with Appveyor

* Remove unwanted space

* Updating Travis script to use JDK 10

* Testing without addons

* Update script for Jacoco report to build 43 profile

* Minor fix in formatting to avoid conflicts

* moving driver specific functions

for SQLServerPreparedStatement

* Remove unwanted code + Update Adal4J library dependency

* changes for CallableStatement

repeptitive delcarations

* Remove an extra bracket due to conflict

* changes for ISQLServerConnection

there are problems with moving all Driver sepcific public methods. SQLServerConnectionPoolProxy also implements this interface and there are many public APIs (such as preparedstmt cacheing stuff) which it doesn't implement, and cannot be moved into the interface at this time.

* lambda touch-up

should generally stick to 1 line if possible.

* changes for ISQLServerDataSource

* updates for ISQLServerResultSet

* Improvements | Missing interface APIs added + Code improvements

* More changes for Interface missing methods

* Implemented missing methods in SQLServerConnectionPoolProxy

* Removed ISQLServerConnection43 for duplicated method definitions

* Added APIs in interface for SQLServerResultSet

* More cleanup done

* Fix minor issues

* Fix test failures and implement Serialization for HashKey

* Fix JavaDoc errors and warnigs

* More changes for CallableStatement APIs

* More changes for Statement and Prepared Statament public APIs

* Javadoc fix

* More changes for SQL Server Bulk Record interface

* Callable Statement missing APIs for Interface

* Add missing desciptions

* Reverting pom.xml change for this PR

* Attempt to resolve conflicts

* Remove Interface as not needed.

* Added missing docs

* Changes for Clob/Blob classes for compliance

* Update ADAL4J with latest version

* Changes for Data Source classes

* Minor fixes to the new changes

* Fix for failing tests

* More changes for compliance

* Add Javadocs and class headers

* Fixed Malformed HTML Error in Javadocs

* javadoc changes

* more javadoc changes to make the abbreviations more clear

* fix unchecked warning issue

* Change HashKey in the driver to 256 Hash

* Add Interface back to SQLServerConnection43 class

* Revert "Change HashKey in the driver to 256 Hash"

This reverts commit e6bef4e.

* Changes for exceptions to throw SQLServerException type

* 6.5.4 preview release changelog (#731)

Release | Changelog for 6.5.4 preview release (#731)

* Fix Conflict issues with master branch
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants