| 1 |
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> |
| 2 |
<!-- |
| 3 |
Copyright 2004-2019 H2 Group. Multiple-Licensed under the MPL 2.0, |
| 4 |
and the EPL 1.0 (http://h2database.com/html/license.html). |
| 5 |
Initial Developer: H2 Group |
| 6 |
--> |
| 7 |
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"> |
| 8 |
<head> |
| 9 |
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" /> |
| 10 |
<meta name="viewport" content="width=device-width, initial-scale=1" /> |
| 11 |
<title> |
| 12 |
Roadmap |
| 13 |
</title> |
| 14 |
<link rel="stylesheet" type="text/css" href="stylesheet.css" /> |
| 15 |
<!-- [search] { --> |
| 16 |
<script type="text/javascript" src="navigation.js"></script> |
| 17 |
</head><body onload="frameMe();"> |
| 18 |
<table class="content"><tr class="content"><td class="content"><div class="contentDiv"> |
| 19 |
<!-- } --> |
| 20 |
|
| 21 |
<h1>Roadmap</h1> |
| 22 |
<p> |
| 23 |
New (feature) requests will usually be added at the very end of the list. The priority is increased for important and popular requests. |
| 24 |
Of course, patches are always welcome, but are not always applied as is. |
| 25 |
See also <a href="build.html#providing_patches">Providing Patches</a>. |
| 26 |
</p> |
| 27 |
|
| 28 |
<h2>Version 1.5.x: Planned Changes</h2> |
| 29 |
<ul><li>Replace file password hash with file encryption key; validate encryption key when connecting. |
| 30 |
</li><li>Remove "set binary collation" feature. |
| 31 |
</li><li>Remove the encryption algorithm XTEA. |
| 32 |
</li><li>Disallow referencing other tables in a table (via constraints for example). |
| 33 |
</li><li>Remove PageStore features like compress_lob. |
| 34 |
</li></ul> |
| 35 |
|
| 36 |
<h2>Version 1.4.x: Planned Changes</h2> |
| 37 |
<ul><li>Change license to MPL 2.0. |
| 38 |
</li><li>Automatic migration from 1.3 databases to 1.4. |
| 39 |
</li><li>Option to disable the file name suffix somehow (issue 447). |
| 40 |
</li></ul> |
| 41 |
|
| 42 |
<h2>Priority 1</h2> |
| 43 |
<ul><li>Bugfixes. |
| 44 |
</li><li>Server side cursors. |
| 45 |
</li></ul> |
| 46 |
|
| 47 |
<h2>Priority 2</h2> |
| 48 |
<ul><li>Support hints for the optimizer (which index to use, enforce the join order). |
| 49 |
</li><li>Full outer joins. |
| 50 |
</li><li>Access rights: remember the owner of an object. |
| 51 |
Create, alter and drop privileges. |
| 52 |
COMMENT: allow owner of object to change it. |
| 53 |
Issue 208: Access rights for schemas. |
| 54 |
</li><li>Test multi-threaded in-memory db access. |
| 55 |
</li><li>Support GRANT SELECT, UPDATE ON [schemaName.] *. |
| 56 |
</li><li>Migrate database tool (also from other database engines). For Oracle, maybe use |
| 57 |
DBMS_METADATA.GET_DDL / GET_DEPENDENT_DDL. |
| 58 |
</li><li>Clustering: support mixed clustering mode (one embedded, others in server mode). |
| 59 |
</li><li>Clustering: reads should be randomly distributed (optional) or to a designated database on RAM (parameter: READ_FROM=3). |
| 60 |
</li><li>PostgreSQL catalog: use BEFORE SELECT triggers instead of views over metadata tables. |
| 61 |
</li><li>Test very large databases and LOBs (up to 256 GB). |
| 62 |
</li><li>Store all temp files in the temp directory. |
| 63 |
</li><li>Make DDL (Data Definition) operations transactional. |
| 64 |
</li><li>Deferred integrity checking (DEFERRABLE INITIALLY DEFERRED). |
| 65 |
</li><li>Groovy Stored Procedures: http://groovy.codehaus.org/GSQL |
| 66 |
</li><li>Add a migration guide (list differences between databases). |
| 67 |
</li><li>Optimization: automatic index creation suggestion using the trace file? |
| 68 |
</li><li>Fulltext search Lucene: analyzer configuration, mergeFactor. |
| 69 |
</li><li>Compression performance: don't allocate buffers, compress / expand in to out buffer. |
| 70 |
</li><li>Rebuild index functionality to shrink index size and improve performance. |
| 71 |
</li><li>Console: add accesskey to most important commands (A, AREA, BUTTON, INPUT, LABEL, LEGEND, TEXTAREA). |
| 72 |
</li><li>Test performance again with SQL Server, Oracle, DB2. |
| 73 |
</li><li>Test with Spatial DB in a box / JTS: http://www.opengeospatial.org/standards/sfs - OpenGIS Implementation Specification. |
| 74 |
</li><li>Improve documentation of MVCC (Multi Version Concurrency Control). |
| 75 |
</li><li>Find a tool to view large text file (larger than 100 MB), with find, page up and down (like less), truncate before / after. |
| 76 |
</li><li>Implement, test, document XAConnection and so on. |
| 77 |
</li><li>Pluggable data type (for streaming, hashing, compression, validation, conversion, encryption). |
| 78 |
</li><li>CHECK: find out what makes CHECK=TRUE slow, move to assertions. |
| 79 |
</li><li>Drop with invalidate views (so that source code is not lost). Check what other databases do exactly. |
| 80 |
</li><li>Index usage for (ID, NAME)=(1, 'Hi'); document. |
| 81 |
</li><li>Set a connection read only (Connection.setReadOnly) or using a connection parameter. |
| 82 |
</li><li>Access rights: finer grained access control (grant access for specific functions). |
| 83 |
</li><li>Version check: docs / web console (using Javascript), and maybe in the library (using TCP/IP). |
| 84 |
</li><li>Web server classloader: override findResource / getResourceFrom. |
| 85 |
</li><li>Cost for embedded temporary view is calculated wrong, if result is constant. |
| 86 |
</li><li>Count index range query (count(*) where id between 10 and 20). |
| 87 |
</li><li>Performance: update in-place. |
| 88 |
</li><li>Clustering: when a database is back alive, automatically synchronize with the master (requires readable transaction log). |
| 89 |
</li><li>Database file name suffix: a way to use no or a different suffix (for example using a slash). |
| 90 |
</li><li>Eclipse plugin. |
| 91 |
</li><li>Asynchronous queries to support publish/subscribe: SELECT ... FOR READ WAIT [maxMillisToWait]. |
| 92 |
See also MS SQL Server "Query Notification". |
| 93 |
</li><li>Fulltext search (native): reader / tokenizer / filter. |
| 94 |
</li><li>Linked schema using CSV files: one schema for a directory of files; support indexes for CSV files. |
| 95 |
</li><li>iReport to support H2. |
| 96 |
</li><li>Include SMTP (mail) client (alert on cluster failure, low disk space,...). |
| 97 |
</li><li>Option for SCRIPT to append to a file. |
| 98 |
</li><li>JSON parser and functions. |
| 99 |
</li><li>Copy database: tool with config GUI and batch mode, extensible (example: compare). |
| 100 |
</li><li>Document, implement tool for long running transactions using user-defined compensation statements. |
| 101 |
</li><li>Support SET TABLE DUAL READONLY. |
| 102 |
</li><li>Events for: database Startup, Connections, Login attempts, Disconnections, Prepare (after parsing), Web Server. See http://docs.openlinksw.com/virtuoso/fn_dbev_startup.html |
| 103 |
</li><li>Optimization: simpler log compression. |
| 104 |
</li><li>Support more standard INFORMATION_SCHEMA tables, as defined in SQL standard. |
| 105 |
</li><li>Compatibility: in MySQL, HSQLDB, /0.0 is NULL; in PostgreSQL, Derby: division by zero. HSQLDB: 0.0e1 / 0.0e1 is NaN. |
| 106 |
</li><li>Functional tables should accept parameters from other tables (see FunctionMultiReturn) SELECT * FROM TEST T, P2C(T.A, T.R). |
| 107 |
</li><li>Custom class loader to reload functions on demand. |
| 108 |
</li><li>Test http://mysql-je.sourceforge.net/ |
| 109 |
</li><li>H2 Console: the webclient could support more features like phpMyAdmin. |
| 110 |
</li><li>Support Oracle functions: TO_NUMBER. |
| 111 |
</li><li>Work on the Java to C converter. |
| 112 |
</li><li>The HELP information schema can be directly exposed in the Console. |
| 113 |
</li><li>Maybe use the 0x1234 notation for binary fields, see MS SQL Server. |
| 114 |
</li><li>Support Oracle CONNECT BY in some way: http://www.adp-gmbh.ch/ora/sql/connect_by.html http://philip.greenspun.com/sql/trees.html |
| 115 |
</li><li>SQL 2003: http://www.wiscorp.com/sql_2003_standard.zip |
| 116 |
</li><li>Version column (number/sequence and timestamp based). |
| 117 |
</li><li>Test and document UPDATE TEST SET (ID, NAME) = (SELECT ID*10, NAME || '!' FROM TEST T WHERE T.ID=TEST.ID). |
| 118 |
</li><li>Max memory rows / max undo log size: use block count / row size not row count. |
| 119 |
</li><li>Implement point-in-time recovery. |
| 120 |
</li><li>Support PL/SQL (programming language / control flow statements). |
| 121 |
</li><li>LIKE: improved version for larger texts (currently using naive search). |
| 122 |
</li><li>Throw an exception when the application calls getInt on a Long (optional). |
| 123 |
</li><li>Default date format for input and output (local date constants). |
| 124 |
</li><li>Document ROWNUM usage for reports: SELECT ROWNUM, * FROM (subquery). |
| 125 |
</li><li>File system that writes to two file systems (replication, replicating file system). |
| 126 |
</li><li>Standalone tool to get relevant system properties and add it to the trace output. |
| 127 |
</li><li>Support 'call proc(1=value)' (PostgreSQL, Oracle). |
| 128 |
</li><li>Console: improve editing data (Tab, Shift-Tab, Enter, Up, Down, Shift+Del?). |
| 129 |
</li><li>Console: autocomplete Ctrl+Space inserts template. |
| 130 |
</li><li>Option to encrypt .trace.db file. |
| 131 |
</li><li>Auto-Update feature for database, .jar file. |
| 132 |
</li><li>ResultSet SimpleResultSet.readFromURL(String url): id varchar, state varchar, released timestamp. |
| 133 |
</li><li>Partial indexing (see PostgreSQL). |
| 134 |
</li><li>Add GUI to build a custom version (embedded, fulltext,...) using build flags. |
| 135 |
</li><li>http://rubyforge.org/projects/hypersonic/ |
| 136 |
</li><li>Add a sample application that runs the H2 unit test and writes the result to a file (so it can be included in the user app). |
| 137 |
</li><li>Table order: ALTER TABLE TEST ORDER BY NAME DESC (MySQL compatibility). |
| 138 |
</li><li>Backup tool should work with other databases as well. |
| 139 |
</li><li>Console: -ifExists doesn't work for the console. Add a flag to disable other dbs. |
| 140 |
</li><li>Check if 'FSUTIL behavior set disablelastaccess 1' improves the performance (fsutil behavior query disablelastaccess). |
| 141 |
</li><li>Java static code analysis: https://pmd.github.io/ |
| 142 |
</li><li>Compatibility for CREATE SCHEMA AUTHORIZATION. |
| 143 |
</li><li>Implement Clob / Blob truncate and the remaining functionality. |
| 144 |
</li><li>File locking: writing a system property to detect concurrent access from the same VM (different classloaders). |
| 145 |
</li><li>Pure SQL triggers (example: update parent table if the child table is changed). |
| 146 |
</li><li>Add H2 to Gem (Ruby install system). |
| 147 |
</li><li>Support linked JCR tables. |
| 148 |
</li><li>Native fulltext search: min word length; store word positions. |
| 149 |
</li><li>Add an option to the SCRIPT command to generate only portable / standard SQL. |
| 150 |
</li><li>Updatable views: create 'instead of' triggers automatically if possible (simple cases first). |
| 151 |
</li><li>Improve create index performance. |
| 152 |
</li><li>Compact databases without having to close the database (vacuum). |
| 153 |
</li><li>Implement more JDBC 4.0 features. |
| 154 |
</li><li>Support TRANSFORM / PIVOT as in MS Access. |
| 155 |
</li><li>Support updatable views with join on primary keys (to extend a table). |
| 156 |
</li><li>Public interface for functions (not public static). |
| 157 |
</li><li>Support reading the transaction log. |
| 158 |
</li><li>Feature matrix. |
| 159 |
</li><li>Updatable result set on table without primary key or unique index. |
| 160 |
</li><li>Allow execution time prepare for SELECT * FROM CSVREAD(?, 'columnNameString') |
| 161 |
</li><li>Support nested transactions (possibly using savepoints internally). |
| 162 |
</li><li>Add a benchmark for bigger databases, and one for many users. |
| 163 |
</li><li>Compression in the result set over TCP/IP. |
| 164 |
</li><li>Support curtimestamp (like curtime, curdate). |
| 165 |
</li><li>Support ANALYZE {TABLE|INDEX} tableName COMPUTE|ESTIMATE|DELETE STATISTICS ptnOption options. |
| 166 |
</li><li>Release locks (shared or exclusive) on demand |
| 167 |
</li><li>Support OUTER UNION |
| 168 |
</li><li>Support parameterized views (similar to CSVREAD, but using just SQL for the definition) |
| 169 |
</li><li>A way (JDBC driver) to map an URL (jdbc:h2map:c1) to a connection object |
| 170 |
</li><li>Support dynamic linked schema (automatically adding/updating/removing tables) |
| 171 |
</li><li>Clustering: adding a node should be very fast and without interrupting clients (very short lock) |
| 172 |
</li><li>Compatibility: # is the start of a single line comment (MySQL) but date quote (Access). Mode specific |
| 173 |
</li><li>Run benchmarks with Android, Java 7, java -server |
| 174 |
</li><li>Optimizations: faster hash function for strings. |
| 175 |
</li><li>DatabaseEventListener: callback for all operations (including expected time, RUNSCRIPT) and cancel functionality |
| 176 |
</li><li>Benchmark: add a graph to show how databases scale (performance/database size) |
| 177 |
</li><li>Implement a SQLData interface to map your data over to a custom object |
| 178 |
</li><li>In the MySQL and PostgreSQL mode, use lower case identifiers by default (DatabaseMetaData.storesLowerCaseIdentifiers = true) |
| 179 |
</li><li>Support multiple directories (on different hard drives) for the same database |
| 180 |
</li><li>Server protocol: use challenge response authentication, but client sends hash(user+password) encrypted with response |
| 181 |
</li><li>Support EXEC[UTE] (doesn't return a result set, compatible to MS SQL Server) |
| 182 |
</li><li>Support native XML data type - see http://en.wikipedia.org/wiki/SQL/XML |
| 183 |
</li><li>Support triggers with a string property or option: SpringTrigger, OSGITrigger |
| 184 |
</li><li>MySQL compatibility: update test1 t1, test2 t2 set t1.id = t2.id where t1.id = t2.id; |
| 185 |
</li><li>Ability to resize the cache array when resizing the cache |
| 186 |
</li><li>Time based cache writing (one second after writing the log) |
| 187 |
</li><li>Check state of H2 driver for DDLUtils: http://issues.apache.org/jira/browse/DDLUTILS-185 |
| 188 |
</li><li>Index usage for REGEXP LIKE. |
| 189 |
</li><li>Compatibility: add a role DBA (like ADMIN). |
| 190 |
</li><li>Better support multiple processors for in-memory databases. |
| 191 |
</li><li>Support N'text' |
| 192 |
</li><li>Support compatibility for jdbc:hsqldb:res: |
| 193 |
</li><li>HSQLDB compatibility: automatically convert to the next 'higher' data type. |
| 194 |
Example: cast(2000000000 as int) + cast(2000000000 as int); |
| 195 |
(HSQLDB: long; PostgreSQL: integer out of range) |
| 196 |
</li><li>Provide an Java SQL builder with standard and H2 syntax |
| 197 |
</li><li>Trace: write OS, file system, JVM,... when opening the database |
| 198 |
</li><li>Support indexes for views (probably requires materialized views) |
| 199 |
</li><li>Document SET SEARCH_PATH, BEGIN, EXECUTE, parameters |
| 200 |
</li><li>Server: use one listener (detect if the request comes from an PG or TCP client) |
| 201 |
</li><li>Optimize SELECT MIN(ID), MAX(ID), COUNT(*) FROM TEST WHERE ID BETWEEN 100 AND 200 |
| 202 |
</li><li>Sequence: PostgreSQL compatibility (rename, create) http://www.postgresql.org/docs/8.2/static/sql-altersequence.html |
| 203 |
</li><li>Support a special trigger on all tables to allow building a transaction log reader. |
| 204 |
</li><li>File system with a background writer thread; test if this is faster |
| 205 |
</li><li>Better document the source code (high level documentation). |
| 206 |
</li><li>Support select * from dual a left join dual b on b.x=(select max(x) from dual) |
| 207 |
</li><li>Optimization: don't lock when the database is read-only |
| 208 |
</li><li>Issue 146: Support merge join. |
| 209 |
</li><li>Integrate spatial functions from http://geosysin.iict.ch/irstv-trac/wiki/H2spatial/Download |
| 210 |
</li><li>Cluster: hot deploy (adding a node at runtime). |
| 211 |
</li><li>Support DatabaseMetaData.insertsAreDetected: updatable result sets should detect inserts. |
| 212 |
</li><li>Native search: support "phrase search", wildcard search (* and ?), case-insensitive search, boolean operators, and grouping |
| 213 |
</li><li>Improve documentation of access rights. |
| 214 |
</li><li>Support opening a database that is in the classpath, maybe using a new file system. Workaround: detect jar file using getClass().getProtectionDomain().getCodeSource().getLocation(). |
| 215 |
</li><li>Remember the user defined data type (domain) of a column. |
| 216 |
</li><li>Auto-server: add option to define the port range or list. |
| 217 |
</li><li>Support Jackcess (MS Access databases) |
| 218 |
</li><li>Built-in methods to write large objects (BLOB and CLOB): FILE_WRITE('test.txt', 'Hello World') |
| 219 |
</li><li>Improve time to open large databases (see mail 'init time for distributed setup') |
| 220 |
</li><li>Move Maven 2 repository from hsql.sf.net to h2database.sf.net |
| 221 |
</li><li>Java 1.5 tool: JdbcUtils.closeSilently(s1, s2,...) |
| 222 |
</li><li>Optimize A=? OR B=? to UNION if the cost is lower. |
| 223 |
</li><li>Javadoc: document design patterns used |
| 224 |
</li><li>Support custom collators, for example for natural sort (for text that contains numbers). |
| 225 |
</li><li>Write an article about SQLInjection (h2/src/docsrc/html/images/SQLInjection.txt) |
| 226 |
</li><li>Convert SQL-injection-2.txt to html document, include SQLInjection.java sample |
| 227 |
</li><li>Support OUT parameters in user-defined procedures. |
| 228 |
</li><li>Web site design: http://www.igniterealtime.org/projects/openfire/index.jsp |
| 229 |
</li><li>HSQLDB compatibility: Openfire server uses: CREATE SCHEMA PUBLIC AUTHORIZATION DBA; |
| 230 |
CREATE USER SA PASSWORD ""; GRANT DBA TO SA; SET SCHEMA PUBLIC |
| 231 |
</li><li>Translation: use ${.} in help.csv |
| 232 |
</li><li>Translated .pdf |
| 233 |
</li><li>Recovery tool: bad blocks should be converted to INSERT INTO SYSTEM_ERRORS(...), and things should go into the .trace.db file |
| 234 |
</li><li>RECOVER=2 to backup the database, run recovery, open the database |
| 235 |
</li><li>Recovery should work with encrypted databases |
| 236 |
</li><li>Corruption: new error code, add help |
| 237 |
</li><li>Space reuse: after init, scan all storages and free those that don't belong to a live database object |
| 238 |
</li><li>Access rights: add missing features (users should be 'owner' of objects; missing rights for sequences; dropping objects) |
| 239 |
</li><li>Support NOCACHE table option (Oracle). |
| 240 |
</li><li>Support table partitioning. |
| 241 |
</li><li>The database should be kept open for a longer time when using the server mode. |
| 242 |
</li><li>Javadocs: for each tool, add a copy & paste sample in the class level. |
| 243 |
</li><li>Javadocs: add @author tags. |
| 244 |
</li><li>Fluent API for tools: Server.createTcpServer().setPort(9081).setPassword(password).start(); |
| 245 |
</li><li>MySQL compatibility: real SQL statement for DESCRIBE TEST |
| 246 |
</li><li>Use a default delay of 1 second before closing a database. |
| 247 |
</li><li>Write (log) to system table before adding to internal data structures. |
| 248 |
</li><li>Support other array types (String[], double[]) in PreparedStatement.setObject(int, Object) (with test case). |
| 249 |
</li><li>Oracle compatibility: support NLS_DATE_FORMAT. |
| 250 |
</li><li>Support for Thread.interrupt to cancel running statements. |
| 251 |
</li><li>Cluster: add feature to make sure cluster nodes can not get out of sync (for example by stopping one process). |
| 252 |
</li><li>H2 Console: support CLOB/BLOB download using a link. |
| 253 |
</li><li>Support flashback queries as in Oracle. |
| 254 |
</li><li>Import / Export of fixed with text files. |
| 255 |
</li><li>HSQLDB compatibility: automatic data type for SUM if value is the value is too big (by default use the same type as the data). |
| 256 |
</li><li>Improve the optimizer to select the right index for special cases: where id between 2 and 4 and booleanColumn |
| 257 |
</li><li>Linked tables: make hidden columns available (Oracle: rowid and ora_rowscn columns). |
| 258 |
</li><li>H2 Console: in-place autocomplete. |
| 259 |
</li><li>Support large databases: split database files to multiple directories / disks (similar to tablespaces). |
| 260 |
</li><li>H2 Console: support configuration option for fixed width (monospace) font. |
| 261 |
</li><li>Native fulltext search: support analyzers (specially for Chinese, Japanese). |
| 262 |
</li><li>Automatically compact databases from time to time (as a background process). |
| 263 |
</li><li>Test Eclipse DTP. |
| 264 |
</li><li>H2 Console: autocomplete: keep the previous setting |
| 265 |
</li><li>executeBatch: option to stop at the first failed statement. |
| 266 |
</li><li>Implement OLAP features as described here: http://www.devx.com/getHelpOn/10MinuteSolution/16573/0/page/5 |
| 267 |
</li><li>Support Oracle ROWID (unique identifier for each row). |
| 268 |
</li><li>MySQL compatibility: alter table add index i(c), add constraint c foreign key(c) references t(c); |
| 269 |
</li><li>Server mode: improve performance for batch updates. |
| 270 |
</li><li>Applets: support read-only databases in a zip file (accessed as a resource). |
| 271 |
</li><li>Long running queries / errors / trace system table. |
| 272 |
</li><li>Better document FTL_SEARCH, FTL_SEARCH_DATA. |
| 273 |
</li><li>Sequences: CURRVAL should be session specific. Compatibility with PostgreSQL. |
| 274 |
</li><li>Index creation using deterministic functions. |
| 275 |
</li><li>ANALYZE: for unique indexes that allow null, count the number of null. |
| 276 |
</li><li>MySQL compatibility: multi-table delete: DELETE .. FROM .. [,...] USING - See http://dev.mysql.com/doc/refman/5.0/en/delete.html |
| 277 |
</li><li>AUTO_SERVER: support changing IP addresses (disable a network while the database is open). |
| 278 |
</li><li>Avoid using java.util.Calendar internally because it's slow, complicated, and buggy. |
| 279 |
</li><li>Support TRUNCATE .. CASCADE like PostgreSQL. |
| 280 |
</li><li>Fulltext search: lazy result generation using SimpleRowSource. |
| 281 |
</li><li>Fulltext search: support alternative syntax: WHERE FTL_CONTAINS(name, 'hello'). |
| 282 |
</li><li>MySQL compatibility: support INSERT INTO table SET column1 = value1, column2 = value2 |
| 283 |
</li><li>Docs: add a one line description for each functions and SQL statements at the top (in the link section). |
| 284 |
</li><li>Javadoc search: weight for titles should be higher ('random' should list Functions as the best match). |
| 285 |
</li><li>Replace information_schema tables with regular tables that are automatically re-built when needed. Use indexes. |
| 286 |
</li><li>Issue 50: Oracle compatibility: support calling 0-parameters functions without parenthesis. Make constants obsolete. |
| 287 |
</li><li>MySQL, HSQLDB compatibility: support where 'a'=1 (not supported by Derby, PostgreSQL) |
| 288 |
</li><li>Finer granularity for SLF4J trace - See http://code.google.com/p/h2database/issues/detail?id=62 |
| 289 |
</li><li>Add database creation date and time to the database. |
| 290 |
</li><li>Support ASSERTION. |
| 291 |
</li><li>MySQL compatibility: support comparing 1='a' |
| 292 |
</li><li>Support PostgreSQL lock modes: http://www.postgresql.org/docs/8.3/static/explicit-locking.html |
| 293 |
</li><li>PostgreSQL compatibility: test DbVisualizer and Squirrel SQL using a new PostgreSQL JDBC driver. |
| 294 |
</li><li>RunScript should be able to read from system in (or quite mode for Shell). |
| 295 |
</li><li>Natural join: support select x from dual natural join dual. |
| 296 |
</li><li>Support using system properties in database URLs (may be a security problem). |
| 297 |
</li><li>Natural join: somehow support this: select a.x, b.x, x from dual a natural join dual b |
| 298 |
</li><li>Use the Java service provider mechanism to register file systems and function libraries. |
| 299 |
</li><li>MySQL compatibility: for auto_increment columns, convert 0 to next value (as when inserting NULL). |
| 300 |
</li><li>Optimization for multi-column IN: use an index if possible. Example: (A, B) IN((1, 2), (2, 3)). |
| 301 |
</li><li>Optimization for EXISTS: convert to inner join or IN(..) if possible. |
| 302 |
</li><li>Serialized file lock: support long running queries. |
| 303 |
</li><li>Network: use 127.0.0.1 if other addresses don't work. |
| 304 |
</li><li>Pluggable network protocol (currently Socket/ServerSocket over TCP/IP) - see also TransportServer with master slave replication. |
| 305 |
</li><li>Support reading JCR data: one table per node type; query table; cache option |
| 306 |
</li><li>OSGi: create a sample application, test, document. |
| 307 |
</li><li>help.csv: use complete examples for functions; run as test case. |
| 308 |
</li><li>Functions to calculate the memory and disk space usage of a table, a row, or a value. |
| 309 |
</li><li>Re-implement PooledConnection; use a lightweight connection object. |
| 310 |
</li><li>Doclet: convert tests in javadocs to a java class. |
| 311 |
</li><li>Doclet: format fields like methods, but support sorting by name and value. |
| 312 |
</li><li>Doclet: shrink the html files. |
| 313 |
</li><li>Allow to scan index backwards starting with a value (to better support ORDER BY DESC). |
| 314 |
</li><li>Java Service Wrapper: try http://yajsw.sourceforge.net/ |
| 315 |
</li><li>Batch parameter for INSERT, UPDATE, and DELETE, and commit after each batch. See also MySQL DELETE. |
| 316 |
</li><li>Use a lazy and auto-close input stream (open resource when reading, close on eof). |
| 317 |
</li><li>Connection pool: 'reset session' command (delete temp tables, rollback, auto-commit true). |
| 318 |
</li><li>Improve SQL documentation, see http://www.w3schools.com/sql/ |
| 319 |
</li><li>MySQL compatibility: DatabaseMetaData.stores*() methods should return the same values. Test with SquirrelSQL. |
| 320 |
</li><li>Sybase/DB2/Oracle compatibility: support out parameters in stored procedures - See http://code.google.com/p/h2database/issues/detail?id=83 |
| 321 |
</li><li>Combine Server and Console tool (only keep Server). |
| 322 |
</li><li>Store the Lucene index in the database itself. |
| 323 |
</li><li>MVCC: compare concurrent update behavior with PostgreSQL and Oracle. |
| 324 |
</li><li>HSQLDB compatibility: CREATE FUNCTION (maybe using a Function interface). |
| 325 |
</li><li>HSQLDB compatibility: support CALL "java.lang.Math.sqrt"(2.0) |
| 326 |
</li><li>Support comma as the decimal separator in the CSV tool. |
| 327 |
</li><li>Compatibility: Java functions with SQLJ Part1 http://www.acm.org/sigmod/record/issues/9912/standards.pdf.gz |
| 328 |
</li><li>Compatibility: Java functions with SQL/PSM (Persistent Stored Modules) - need to find the documentation. |
| 329 |
</li><li>CACHE_SIZE: automatically use a fraction of Runtime.maxMemory - maybe automatically the second level cache. |
| 330 |
</li><li>PostgreSQL compatibility: when in PG mode, treat BYTEA data like PG. |
| 331 |
</li><li>Support =ANY(array) as in PostgreSQL. See also http://www.postgresql.org/docs/8.0/interactive/arrays.html |
| 332 |
</li><li>IBM DB2 compatibility: support PREVIOUS VALUE FOR sequence. |
| 333 |
</li><li>Compatibility: use different LIKE ESCAPE characters depending on the mode (disable for Derby, HSQLDB, DB2, Oracle, MSSQLServer). |
| 334 |
</li><li>FTP: document the server, including -ftpTask option to execute / kill remote processes |
| 335 |
</li><li>FTP: problems with multithreading? |
| 336 |
</li><li>FTP: implement SFTP / FTPS |
| 337 |
</li><li>FTP: access to a database (.csv for a table, a directory for a schema, a file for a lob, a script.sql file). |
| 338 |
</li><li>Improve database file locking (maybe use native file locking). The current approach seems to be problematic |
| 339 |
if the file system is on a remote share (see Google Group 'Lock file modification time is in the future'). |
| 340 |
</li><li>Document internal features such as BELONGS_TO_TABLE, NULL_TO_DEFAULT, SEQUENCE. |
| 341 |
</li><li>Issue 107: Prefer using the ORDER BY index if LIMIT is used. |
| 342 |
</li><li>An index on (id, name) should be used for a query: select * from t where s=? order by i |
| 343 |
</li><li>Support reading sequences using DatabaseMetaData.getTables(null, null, null, new String[]{"SEQUENCE"}). |
| 344 |
See PostgreSQL. |
| 345 |
</li><li>Add option to enable TCP_NODELAY using Socket.setTcpNoDelay(true). |
| 346 |
</li><li>Maybe disallow = within database names (jdbc:h2:mem:MODE=DB2 means database name MODE=DB2). |
| 347 |
</li><li>Fast alter table add column. |
| 348 |
</li><li>Improve concurrency for in-memory database operations. |
| 349 |
</li><li>Issue 122: Support for connection aliases for remote tcp connections. |
| 350 |
</li><li>Fast scrambling (strong encryption doesn't help if the password is included in the application). |
| 351 |
</li><li>H2 Console: support -webPassword to require a password to access preferences or shutdown. |
| 352 |
</li><li>Issue 126: The index name should be "IDX_" plus the constraint name unless there is a conflict, in which case append a number. |
| 353 |
</li><li>Issue 127: Support activation/deactivation of triggers |
| 354 |
</li><li>Issue 130: Custom log event listeners |
| 355 |
</li><li>Issue 132: Use Java enum trigger type. |
| 356 |
</li><li>Issue 134: IBM DB2 compatibility: session global variables. |
| 357 |
</li><li>Cluster: support load balance with values for each server / auto detect. |
| 358 |
</li><li>FTL_SET_OPTION(keyString, valueString) with key stopWords at first. |
| 359 |
</li><li>Pluggable access control mechanism. |
| 360 |
</li><li>Fulltext search (Lucene): support streaming CLOB data. |
| 361 |
</li><li>Document/example how to create and read an encrypted script file. |
| 362 |
</li><li>Fulltext search (Lucene): only prefix column names with _ if they already start with _. Instead of DATA / QUERY / modified use _DATA, _QUERY, _MODIFIED if possible. |
| 363 |
</li><li>Support a way to create or read compressed encrypted script files using an API. |
| 364 |
</li><li>The network client should better detect if the server is not an H2 server and fail early. |
| 365 |
</li><li>H2 Console: support CLOB/BLOB upload. |
| 366 |
</li><li>Database file lock: detect hibernate / standby / very slow threads (compare system time). |
| 367 |
</li><li>Automatic detection of redundant indexes. |
| 368 |
</li><li>Maybe reject join without "on" (except natural join). |
| 369 |
</li><li>Implement GiST (Generalized Search Tree for Secondary Storage). |
| 370 |
</li><li>Function to read a number of bytes/characters from an BLOB or CLOB. |
| 371 |
</li><li>Issue 156: Support SELECT ? UNION SELECT ?. |
| 372 |
</li><li>Automatic mixed mode: support a port range list (to avoid firewall problems). |
| 373 |
</li><li>Support the pseudo column rowid, oid, _rowid_. |
| 374 |
</li><li>H2 Console / large result sets: stream early instead of keeping a whole result in-memory |
| 375 |
</li><li>Support TRUNCATE for linked tables. |
| 376 |
</li><li>UNION: evaluate INTERSECT before UNION (like most other database except Oracle). |
| 377 |
</li><li>Delay creating the information schema, and share metadata columns. |
| 378 |
</li><li>TCP Server: use a nonce (number used once) to protect unencrypted channels against replay attacks. |
| 379 |
</li><li>Simplify running scripts and recovery: CREATE FORCE USER (overwrites an existing user). |
| 380 |
</li><li>Support CREATE DATABASE LINK (a custom JDBC driver is already supported). |
| 381 |
</li><li>Support large GROUP BY operations. Issue 216. |
| 382 |
</li><li>Issue 163: Allow to create foreign keys on metadata types. |
| 383 |
</li><li>Logback: write a native DBAppender. |
| 384 |
</li><li>Cache size: don't use more cache than what is available. |
| 385 |
</li><li>Allow to defragment at runtime (similar to SHUTDOWN DEFRAG) in a background thread. |
| 386 |
</li><li>Tree index: Instead of an AVL tree, use a general balanced trees or a scapegoat tree. |
| 387 |
</li><li>User defined functions: allow to store the bytecode (of just the class, or the jar file of the extension) in the database. |
| 388 |
</li><li>Compatibility: ResultSet.getObject() on a CLOB (TEXT) should return String for PostgreSQL and MySQL. |
| 389 |
</li><li>Optimizer: WHERE X=? AND Y IN(?), it always uses the index on Y. Should be cost based. |
| 390 |
</li><li>Common Table Expression (CTE) / recursive queries: support parameters. Issue 314. |
| 391 |
</li><li>Oracle compatibility: support INSERT ALL. |
| 392 |
</li><li>Issue 178: Optimizer: index usage when both ascending and descending indexes are available. |
| 393 |
</li><li>Issue 179: Related subqueries in HAVING clause. |
| 394 |
</li><li>IBM DB2 compatibility: NOT NULL WITH DEFAULT. Similar to MySQL Mode.convertInsertNullToZero. |
| 395 |
</li><li>Maybe use a different page layout: keep the data at the head of the page, and ignore the tail |
| 396 |
(don't store / read it). This may increase write / read performance depending on the file system. |
| 397 |
</li><li>Indexes of temporary tables are currently kept in-memory. Is this how it should be? |
| 398 |
</li><li>The Shell tool should support the same built-in commands as the H2 Console. |
| 399 |
</li><li>Maybe use PhantomReference instead of finalize. |
| 400 |
</li><li>Database file name suffix: should only have one dot by default. Example: .h2db |
| 401 |
</li><li>Issue 196: Function based indexes |
| 402 |
</li><li>Fix the disk space leak (killing the process at the exact right moment will increase |
| 403 |
the disk space usage; this space is not re-used). See TestDiskSpaceLeak.java |
| 404 |
</li><li>Allow to access the database over HTTP (possibly using port 80) and a servlet in a REST way. |
| 405 |
</li><li>ODBC: encrypted databases are not supported because the ;CIPHER= can not be set. |
| 406 |
</li><li>Support CLOB and BLOB update, specially conn.createBlob().setBinaryStream(1); |
| 407 |
</li><li>Optimizer: index usage when both ascending and descending indexes are available. Issue 178. |
| 408 |
</li><li>Issue 306: Support schema specific domains. |
| 409 |
</li><li>Triggers: support user defined execution order. Oracle: |
| 410 |
CREATE OR REPLACE TRIGGER TEST_2 BEFORE INSERT |
| 411 |
ON TEST FOR EACH ROW FOLLOWS TEST_1. |
| 412 |
SQL specifies that multiple triggers should be fired in time-of-creation order. |
| 413 |
PostgreSQL uses name order, which was judged to be more convenient. |
| 414 |
Derby: triggers are fired in the order in which they were created. |
| 415 |
</li><li>PostgreSQL compatibility: combine "users" and "roles". See: |
| 416 |
http://www.postgresql.org/docs/8.1/interactive/user-manag.html |
| 417 |
</li><li>Improve documentation of system properties: only list the property names, default values, and description. |
| 418 |
</li><li>Support running totals / cumulative sum using SUM(..) OVER(..). |
| 419 |
</li><li>Improve object memory size calculation. Use constants for known VMs, or use reflection to call java.lang.instrument.Instrumentation.getObjectSize(Object objectToSize) |
| 420 |
</li><li>Triggers: NOT NULL checks should be done after running triggers (Oracle behavior, maybe others). |
| 421 |
</li><li>Common Table Expression (CTE) / recursive queries: support INSERT INTO ... SELECT ... Issue 219. |
| 422 |
</li><li>Common Table Expression (CTE) / recursive queries: support non-recursive queries. Issue 217. |
| 423 |
</li><li>Common Table Expression (CTE) / recursive queries: avoid endless loop. Issue 218. |
| 424 |
</li><li>Common Table Expression (CTE) / recursive queries: support multiple named queries. Issue 220. |
| 425 |
</li><li>Common Table Expression (CTE) / recursive queries: identifier scope may be incorrect. Issue 222. |
| 426 |
</li><li>Log long running transactions (similar to long running statements). |
| 427 |
</li><li>Parameter data type is data type of other operand. Issue 205. |
| 428 |
</li><li>Some combinations of nested join with right outer join are not supported. |
| 429 |
</li><li>DatabaseEventListener.openConnection(id) and closeConnection(id). |
| 430 |
</li><li>Listener or authentication module for new connections, or a way to restrict the number of different connections to a tcp server, |
| 431 |
or to prevent to login with the same username and password from different IPs. |
| 432 |
Possibly using the DatabaseEventListener API, or a new API. |
| 433 |
</li><li>Compatibility for data type CHAR (Derby, HSQLDB). Issue 212. |
| 434 |
</li><li>Optimizer: use a histogram of the data, specially for non-normal distributions. |
| 435 |
</li><li>Trigger: allow declaring as source code (like functions). |
| 436 |
</li><li>User defined aggregate: allow declaring as source code (like functions). |
| 437 |
</li><li>The error "table not found" is sometimes caused by using the wrong database. |
| 438 |
Add "(this database is empty)" to the exception message if applicable. |
| 439 |
</li><li>MySQL + PostgreSQL compatibility: support string literal escape with \n. |
| 440 |
</li><li>PostgreSQL compatibility: support string literal escape with double \\. |
| 441 |
</li><li>Document the TCP server "management_db". Maybe include the IP address of the client. |
| 442 |
</li><li>Use javax.tools.JavaCompilerTool instead of com.sun.tools.javac.Main |
| 443 |
</li><li>If a database object was not found in the current schema, but one with the same name existed in another schema, included that in the error message. |
| 444 |
</li><li>Optimization to use an index for OR when using multiple keys: where (key1 = ? and key2 = ?) OR (key1 = ? and key2 = ?) |
| 445 |
</li><li>Issue 302: Support optimizing queries with both inner and outer joins, as in: |
| 446 |
select * from test a inner join test b on a.id=b.id inner join o on o.id=a.id where b.x=1 |
| 447 |
(the optimizer should swap a and b here). |
| 448 |
See also TestNestedJoins, tag "swapInnerJoinTables". |
| 449 |
</li><li>Move table to a different schema (rename table to a different schema), possibly using ALTER TABLE ... SET SCHEMA ...; |
| 450 |
</li><li>nioMapped file system: automatically fall back to regular (non mapped) IO if there is a problem (out of memory exception for example). |
| 451 |
</li><li>Column as parameter of function table. Issue 228. |
| 452 |
</li><li>Connection pool: detect ;AUTOCOMMIT=FALSE in the database URL, and if set, |
| 453 |
disable autocommit for all connections. |
| 454 |
</li><li>Compatibility with MS Access: support "&" to concatenate text. |
| 455 |
</li><li>The BACKUP statement should not synchronize on the database, and therefore should not block other users. |
| 456 |
</li><li>Document the database file format. |
| 457 |
</li><li>Support reading LOBs. |
| 458 |
</li><li>Require appending DANGEROUS=TRUE when using certain dangerous settings such as |
| 459 |
LOG=0, LOG=1, LOCK_MODE=0, disabling FILE_LOCK,... |
| 460 |
</li><li>Support UDT (user defined types) similar to how Apache Derby supports it: |
| 461 |
check constraint, allow to use it in Java functions as parameters (return values already seem to work). |
| 462 |
</li><li>Encrypted file system (use cipher text stealing so file length doesn't need to decrypt; 4 KB header per file, |
| 463 |
optional compatibility with current encrypted database files). |
| 464 |
</li><li>Issue 229: SELECT with simple OR tests uses tableScan when it could use indexes. |
| 465 |
</li><li>GROUP BY queries should use a temporary table if there are too many rows. |
| 466 |
</li><li>BLOB: support random access when reading. |
| 467 |
</li><li>CLOB: support random access when reading (this is harder than for BLOB as data is stored in UTF-8 form). |
| 468 |
</li><li>Compatibility: support SELECT INTO (as an alias for CREATE TABLE ... AS SELECT ...). |
| 469 |
</li><li>Compatibility with MySQL: support SELECT INTO OUTFILE (cannot be an existing file) as an alias for CSVWRITE(...). |
| 470 |
</li><li>Compatibility with MySQL: support non-strict mode (sql_mode = "") any data |
| 471 |
that is too large for the column will just be truncated or set to the default value. |
| 472 |
</li><li>The full condition should be sent to the linked table, not just the indexed condition. |
| 473 |
Example: TestLinkedTableFullCondition |
| 474 |
</li><li>Compatibility with IBM DB2: CREATE PROCEDURE. |
| 475 |
</li><li>Compatibility with IBM DB2: SQL cursors. |
| 476 |
</li><li>Single-column primary key values are always stored explicitly. This is not required. |
| 477 |
</li><li>Compatibility with MySQL: support CREATE TABLE TEST(NAME VARCHAR(255) CHARACTER SET UTF8). |
| 478 |
</li><li>CALL is incompatible with other databases because it returns a result set, so that CallableStatement.execute() returns true. |
| 479 |
</li><li>Compatibility for ARRAY data type (Oracle: VARRAY(n) of VARCHAR(m); HSQLDB: VARCHAR(n) ARRAY; Postgres: VARCHAR(n)[]). |
| 480 |
</li><li>PostgreSQL compatible array literal syntax: ARRAY[['a', 'b'], ['c', 'd']] |
| 481 |
</li><li>PostgreSQL compatibility: UPDATE with FROM. |
| 482 |
</li><li>Issue 297: Oracle compatibility for "at time zone". |
| 483 |
</li><li>IBM DB2 compatibility: IDENTITY_VAL_LOCAL(). |
| 484 |
</li><li>Support SQL/XML data type. |
| 485 |
</li><li>Support concurrent opening of databases. |
| 486 |
</li><li>Improved error message and diagnostics in case of network configuration problems. |
| 487 |
</li><li>TRUNCATE should reset the identity columns by default in MySQL and MS SQL Server compatibility modes (and possibly other). |
| 488 |
</li><li>Adding a primary key should make the columns 'not null' unless if there is a row with null |
| 489 |
(compatibility with MySQL, PostgreSQL, HSQLDB; not Derby). |
| 490 |
</li><li>ARRAY data type: support Integer[] and so on in Java functions (currently only Object[] is supported). |
| 491 |
</li><li>MySQL compatibility: LOCK TABLES a READ, b READ - see also http://dev.mysql.com/doc/refman/5.0/en/lock-tables.html |
| 492 |
</li><li>The HTML to PDF converter should use http://code.google.com/p/wkhtmltopdf/ |
| 493 |
</li><li>Issue 303: automatically convert "X NOT IN(SELECT...)" to "NOT EXISTS(...)". |
| 494 |
</li><li>MySQL compatibility: update test1 t1, test2 t2 set t1.name=t2.name where t1.id=t2.id. |
| 495 |
</li><li>Issue 283: Improve performance of H2 on Android. |
| 496 |
</li><li>Support INSERT INTO / UPDATE / MERGE ... RETURNING to retrieve the generated key(s). |
| 497 |
</li><li>Column compression option - see http://groups.google.com/group/h2-database/browse_thread/thread/3e223504e52671fa/243da82244343f5d |
| 498 |
</li><li>MS SQL Server compatibility: support @@ROWCOUNT. |
| 499 |
</li><li>Issue 311: Serialized lock mode: executeQuery of write operations fails. |
| 500 |
</li><li>PostgreSQL compatibility: support PgAdmin III (specially the function current_setting). |
| 501 |
</li><li>Support SELECT ... FOR UPDATE OF [field-list] (supported by PostgreSQL, MySQL, and HSQLDB; but not Derby). |
| 502 |
</li><li>Support SELECT ... FOR UPDATE OF [table-list] (supported by PostgreSQL, HSQLDB, Sybase). |
| 503 |
</li><li>TRANSACTION_ID() for in-memory databases. |
| 504 |
</li><li>TRANSACTION_ID() should be long (same as HSQLDB and PostgreSQL). |
| 505 |
</li><li>Support [INNER | OUTER] JOIN USING(column [,...]). |
| 506 |
</li><li>Support NATURAL [ { LEFT | RIGHT } [ OUTER ] | INNER ] JOIN (Derby, Oracle) |
| 507 |
</li><li>GROUP BY columnNumber (similar to ORDER BY columnNumber) (MySQL, PostgreSQL, SQLite; not by HSQLDB and Derby). |
| 508 |
</li><li>Index conditions: WHERE AGE>1 should not scan through all rows with AGE=1. |
| 509 |
</li><li>PHP support: H2 should support PDO, or test with PostgreSQL PDO. |
| 510 |
</li><li>Outer joins: if no column of the outer join table is referenced, the outer join table could be removed from the query. |
| 511 |
</li><li>Cluster: allow using auto-increment and identity columns by ensuring executed in lock-step. |
| 512 |
</li><li>MySQL compatibility: index names only need to be unique for the given table. |
| 513 |
</li><li>Issue 352: constraints: distinguish between 'no action' and 'restrict'. Currently, only restrict is supported, |
| 514 |
and 'no action' is internally mapped to 'restrict'. The database meta data returns 'restrict' in all cases. |
| 515 |
</li><li>Issue 348: Oracle compatibility: division should return a decimal result. |
| 516 |
</li><li>Read rows on demand: instead of reading the whole row, only read up to that column that is requested. |
| 517 |
Keep an pointer to the data area and the column id that is already read. |
| 518 |
</li><li>Long running transactions: log session id when detected. |
| 519 |
</li><li>Optimization: "select id from test" should use the index on id even without "order by". |
| 520 |
</li><li>Sybase SQL Anywhere compatibility: SELECT TOP ... START AT ... |
| 521 |
</li><li>Issue 390: RUNSCRIPT FROM '...' CONTINUE_ON_ERROR |
| 522 |
</li></ul> |
| 523 |
|
| 524 |
<h2>Not Planned</h2> |
| 525 |
<ul> |
| 526 |
<li>HSQLDB (did) support this: select id i from test where i<0 (other databases don't). Supporting it may break compatibility. |
| 527 |
</li><li>String.intern (so that Strings can be compared with ==) will not be used because some VMs have problems when used extensively. |
| 528 |
</li><li>In prepared statements, identifier names (table names and so on) can not be parameterized. Adding such a feature would complicate the source code without providing reasonable speedup, and would slow down regular prepared statements. |
| 529 |
</li></ul> |
| 530 |
|
| 531 |
<!-- [close] { --></div></td></tr></table><!-- } --><!-- analytics --></body></html> |
| 532 |
|