You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by ty...@apache.org on 2014/05/23 18:40:25 UTC
svn commit: r1597123 - /cassandra/site/publish/doc/cql3/CQL-2.0.html
Author: tylerhobbs
Date: Fri May 23 16:40:25 2014
New Revision: 1597123
URL: http://svn.apache.org/r1597123
Log:
Update CQL docs for CASSANDRA-6875
Modified:
cassandra/site/publish/doc/cql3/CQL-2.0.html
Modified: cassandra/site/publish/doc/cql3/CQL-2.0.html
URL: http://svn.apache.org/viewvc/cassandra/site/publish/doc/cql3/CQL-2.0.html?rev=1597123&r1=1597122&r2=1597123&view=diff
==============================================================================
--- cassandra/site/publish/doc/cql3/CQL-2.0.html (original)
+++ cassandra/site/publish/doc/cql3/CQL-2.0.html Fri May 23 16:40:25 2014
@@ -1,4 +1,4 @@
-<?xml version='1.0' encoding='utf-8' ?><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"/><title>CQL</title></head><body><p><link rel="StyleSheet" href="CQL.css" type="text/css" media="screen"></p><h1 id="CassandraQueryLanguageCQLv3.1.5">Cassandra Query Language (CQL) v3.1.5</h1><span id="tableOfContents"><ol style="list-style: none;"><li><a href="CQL.html#CassandraQueryLanguageCQLv3.1.5">Cassandra Query Language (CQL) v3.1.5</a><ol style="list-style: none;"><li><a href="CQL.html#CQLSyntax">CQL Syntax</a><ol style="list-style: none;"><li><a href="CQL.html#Preamble">Preamble</a></li><li><a href="CQL.html#Conventions">Conventions</a></li><li><a href="CQL.html#identifiers">Identifiers and keywords</a></li><li><a href="CQL.html#constants">Constants</a></li><li><a href="CQL.html#Comments">Comments</a></l
i><li><a href="CQL.html#statements">Statements</a></li><li><a href="CQL.html#preparedStatement">Prepared Statement</a></li></ol></li><li><a href="CQL.html#dataDefinition">Data Definition</a><ol style="list-style: none;"><li><a href="CQL.html#createKeyspaceStmt">CREATE KEYSPACE</a></li><li><a href="CQL.html#useStmt">USE</a></li><li><a href="CQL.html#alterKeyspaceStmt">ALTER KEYSPACE</a></li><li><a href="CQL.html#dropKeyspaceStmt">DROP KEYSPACE</a></li><li><a href="CQL.html#createTableStmt">CREATE TABLE</a></li><li><a href="CQL.html#alterTableStmt">ALTER TABLE</a></li><li><a href="CQL.html#dropTableStmt">DROP TABLE</a></li><li><a href="CQL.html#truncateStmt">TRUNCATE</a></li><li><a href="CQL.html#createIndexStmt">CREATE INDEX</a></li><li><a href="CQL.html#dropIndexStmt">DROP INDEX</a></li></ol></li><li><a href="CQL.html#dataManipulation">Data Manipulation</a><ol style="list-style: none;"><li><a href="CQL.html#insertStmt">INSERT</a></li><li><a href="CQL.html#updateStmt">UPDATE</a></li>
<li><a href="CQL.html#deleteStmt">DELETE</a></li><li><a href="CQL.html#batchStmt">BATCH</a></li></ol></li><li><a href="CQL.html#queries">Queries</a><ol style="list-style: none;"><li><a href="CQL.html#selectStmt">SELECT</a></li></ol></li><li><a href="CQL.html#types">Data Types</a><ol style="list-style: none;"><li><a href="CQL.html#usingdates">Working with dates</a></li><li><a href="CQL.html#counters">Counters</a></li><li><a href="CQL.html#collections">Working with collections</a></li></ol></li><li><a href="CQL.html#functions">Functions</a><ol style="list-style: none;"><li><a href="CQL.html#tokenFun">Token</a></li><li><a href="CQL.html#uuidFun">Uuid</a></li><li><a href="CQL.html#timeuuidFun">Timeuuid functions</a></li><li><a href="CQL.html#blobFun">Blob conversion functions</a></li></ol></li><li><a href="CQL.html#appendixA">Appendix A: CQL Keywords</a></li><li><a href="CQL.html#changes">Changes</a><ol style="list-style: none;"><li><a href="CQL.html#a3.1.5">3.1.5</a></li><li><a href="C
QL.html#a3.1.4">3.1.4</a></li><li><a href="CQL.html#a3.1.3">3.1.3</a></li><li><a href="CQL.html#a3.1.2">3.1.2</a></li><li><a href="CQL.html#a3.1.1">3.1.1</a></li><li><a href="CQL.html#a3.1.0">3.1.0</a></li><li><a href="CQL.html#a3.0.5">3.0.5</a></li><li><a href="CQL.html#a3.0.4">3.0.4</a></li><li><a href="CQL.html#a3.0.3">3.0.3</a></li><li><a href="CQL.html#a3.0.2">3.0.2</a></li><li><a href="CQL.html#a3.0.1">3.0.1</a></li></ol></li><li><a href="CQL.html#Versioning">Versioning</a></li></ol></li></ol></span><h2 id="CQLSyntax">CQL Syntax</h2><h3 id="Preamble">Preamble</h3><p>This document describes the Cassandra Query Language (CQL) version 3. CQL v3 is not backward compatible with CQL v2 and differs from it in numerous ways. Note that this document describes the last version of the languages. However, the <a href="#changes">changes</a> section provides the diff between the different versions of CQL v3.</p><p>CQL v3 offers a model very close to SQL in the sense that data is put in <em>
tables</em> containing <em>rows</em> of <em>columns</em>. For that reason, when used in this document, these terms (tables, rows and columns) have the same definition than they have in SQL. But please note that as such, they do <strong>not</strong> refer to the concept of rows and columns found in the internal implementation of Cassandra and in the thrift and CQL v2 API.</p><h3 id="Conventions">Conventions</h3><p>To aid in specifying the CQL syntax, we will use the following conventions in this document:</p><ul><li>Language rules will be given in a <a href="http://en.wikipedia.org/wiki/Backus%E2%80%93Naur_Form">BNF</a> -like notation:</li></ul><pre class="syntax"><pre><start> ::= TERMINAL <non-terminal1> <non-terminal1>
+<?xml version='1.0' encoding='utf-8' ?><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"/><title>CQL</title></head><body><p><link rel="StyleSheet" href="CQL.css" type="text/css" media="screen"></p><h1 id="CassandraQueryLanguageCQLv3.1.7">Cassandra Query Language (CQL) v3.1.7</h1><span id="tableOfContents"><ol style="list-style: none;"><li><a href="CQL.html#CassandraQueryLanguageCQLv3.1.7">Cassandra Query Language (CQL) v3.1.7</a><ol style="list-style: none;"><li><a href="CQL.html#CQLSyntax">CQL Syntax</a><ol style="list-style: none;"><li><a href="CQL.html#Preamble">Preamble</a></li><li><a href="CQL.html#Conventions">Conventions</a></li><li><a href="CQL.html#identifiers">Identifiers and keywords</a></li><li><a href="CQL.html#constants">Constants</a></li><li><a href="CQL.html#Comments">Comments</a></l
i><li><a href="CQL.html#statements">Statements</a></li><li><a href="CQL.html#preparedStatement">Prepared Statement</a></li></ol></li><li><a href="CQL.html#dataDefinition">Data Definition</a><ol style="list-style: none;"><li><a href="CQL.html#createKeyspaceStmt">CREATE KEYSPACE</a></li><li><a href="CQL.html#useStmt">USE</a></li><li><a href="CQL.html#alterKeyspaceStmt">ALTER KEYSPACE</a></li><li><a href="CQL.html#dropKeyspaceStmt">DROP KEYSPACE</a></li><li><a href="CQL.html#createTableStmt">CREATE TABLE</a></li><li><a href="CQL.html#alterTableStmt">ALTER TABLE</a></li><li><a href="CQL.html#dropTableStmt">DROP TABLE</a></li><li><a href="CQL.html#truncateStmt">TRUNCATE</a></li><li><a href="CQL.html#createIndexStmt">CREATE INDEX</a></li><li><a href="CQL.html#dropIndexStmt">DROP INDEX</a></li></ol></li><li><a href="CQL.html#dataManipulation">Data Manipulation</a><ol style="list-style: none;"><li><a href="CQL.html#insertStmt">INSERT</a></li><li><a href="CQL.html#updateStmt">UPDATE</a></li>
<li><a href="CQL.html#deleteStmt">DELETE</a></li><li><a href="CQL.html#batchStmt">BATCH</a></li></ol></li><li><a href="CQL.html#queries">Queries</a><ol style="list-style: none;"><li><a href="CQL.html#selectStmt">SELECT</a></li></ol></li><li><a href="CQL.html#types">Data Types</a><ol style="list-style: none;"><li><a href="CQL.html#usingdates">Working with dates</a></li><li><a href="CQL.html#counters">Counters</a></li><li><a href="CQL.html#collections">Working with collections</a></li></ol></li><li><a href="CQL.html#functions">Functions</a><ol style="list-style: none;"><li><a href="CQL.html#tokenFun">Token</a></li><li><a href="CQL.html#uuidFun">Uuid</a></li><li><a href="CQL.html#timeuuidFun">Timeuuid functions</a></li><li><a href="CQL.html#blobFun">Blob conversion functions</a></li></ol></li><li><a href="CQL.html#appendixA">Appendix A: CQL Keywords</a></li><li><a href="CQL.html#changes">Changes</a><ol style="list-style: none;"><li><a href="CQL.html#a3.1.7">3.1.7</a></li><li><a href="C
QL.html#a3.1.6">3.1.6</a></li><li><a href="CQL.html#a3.1.5">3.1.5</a></li><li><a href="CQL.html#a3.1.4">3.1.4</a></li><li><a href="CQL.html#a3.1.3">3.1.3</a></li><li><a href="CQL.html#a3.1.2">3.1.2</a></li><li><a href="CQL.html#a3.1.1">3.1.1</a></li><li><a href="CQL.html#a3.1.0">3.1.0</a></li><li><a href="CQL.html#a3.0.5">3.0.5</a></li><li><a href="CQL.html#a3.0.4">3.0.4</a></li><li><a href="CQL.html#a3.0.3">3.0.3</a></li><li><a href="CQL.html#a3.0.2">3.0.2</a></li><li><a href="CQL.html#a3.0.1">3.0.1</a></li></ol></li><li><a href="CQL.html#Versioning">Versioning</a></li></ol></li></ol></span><h2 id="CQLSyntax">CQL Syntax</h2><h3 id="Preamble">Preamble</h3><p>This document describes the Cassandra Query Language (CQL) version 3. CQL v3 is not backward compatible with CQL v2 and differs from it in numerous ways. Note that this document describes the last version of the languages. However, the <a href="#changes">changes</a> section provides the diff between the different versions of CQL
v3.</p><p>CQL v3 offers a model very close to SQL in the sense that data is put in <em>tables</em> containing <em>rows</em> of <em>columns</em>. For that reason, when used in this document, these terms (tables, rows and columns) have the same definition than they have in SQL. But please note that as such, they do <strong>not</strong> refer to the concept of rows and columns found in the internal implementation of Cassandra and in the thrift and CQL v2 API.</p><h3 id="Conventions">Conventions</h3><p>To aid in specifying the CQL syntax, we will use the following conventions in this document:</p><ul><li>Language rules will be given in a <a href="http://en.wikipedia.org/wiki/Backus%E2%80%93Naur_Form">BNF</a> -like notation:</li></ul><pre class="syntax"><pre><start> ::= TERMINAL <non-terminal1> <non-terminal1>
</pre></pre><ul><li>Nonterminal symbols will have <code><angle brackets></code>.</li><li>As additional shortcut notations to BNF, we’ll use traditional regular expression’s symbols (<code>?</code>, <code>+</code> and <code>*</code>) to signify that a given symbol is optional and/or can be repeated. We’ll also allow parentheses to group symbols and the <code>[<characters>]</code> notation to represent any one of <code><characters></code>.</li><li>The grammar is provided for documentation purposes and leave some minor details out. For instance, the last column definition in a <code>CREATE TABLE</code> statement is optional but supported if present even though the provided grammar in this document suggest it is not supported. </li><li>Sample code will be provided in a code block:</li></ul><pre class="sample"><pre>SELECT sample_usage FROM cql;
</pre></pre><ul><li>References to keywords or pieces of CQL code in running text will be shown in a <code>fixed-width font</code>.</li></ul><h3 id="identifiers">Identifiers and keywords</h3><p>The CQL language uses <em>identifiers</em> (or <em>names</em>) to identify tables, columns and other objects. An identifier is a token matching the regular expression <code lang="a-zA-Z">[a-zA-Z0-9_]</code><code>*</code>.</p><p>A number of such identifiers, like <code>SELECT</code> or <code>WITH</code>, are <em>keywords</em>. They have a fixed meaning for the language and most are reserved. The list of those keywords can be found in <a href="#appendixA">Appendix A</a>.</p><p>Identifiers and (unquoted) keywords are case insensitive. Thus <code>SELECT</code> is the same than <code>select</code> or <code>sElEcT</code>, and <code>myId</code> is the same than <code>myid</code> or <code>MYID</code> for instance. A convention often used (in particular by the samples of this documentation) is to use u
pper case for keywords and lower case for other identifiers.</p><p>There is a second kind of identifiers called <em>quoted identifiers</em> defined by enclosing an arbitrary sequence of characters in double-quotes(<code>"</code>). Quoted identifiers are never keywords. Thus <code>"select"</code> is not a reserved keyword and can be used to refer to a column, while <code>select</code> would raise a parse error. Also, contrarily to unquoted identifiers and keywords, quoted identifiers are case sensitive (<code>"My Quoted Id"</code> is <em>different</em> from <code>"my quoted id"</code>). A fully lowercase quoted identifier that matches <code lang="a-zA-Z">[a-zA-Z0-9_]</code><code>*</code> is equivalent to the unquoted identifier obtained by removing the double-quote (so <code>"myid"</code> is equivalent to <code>myid</code> and to <code>myId</code> but different from <code>"myId"</code>). Inside a quoted identifier, the double-quote character can be repeated to escape it, so <code>"fo
o "" bar"</code> is a valid identifier.</p><h3 id="constants">Constants</h3><p>CQL defines the following kind of <em>constants</em>: strings, integers, floats, booleans, uuids and blobs:</p><ul><li>A string constant is an arbitrary sequence of characters characters enclosed by single-quote(<code>'</code>). One can include a single-quote in a string by repeating it, e.g. <code>'It''s raining today'</code>. Those are not to be confused with quoted identifiers that use double-quotes.</li><li>An integer constant is defined by <code>'-'?[0-9]+</code>.</li><li>A float constant is defined by <code>'-'?[0-9]+('.'[0-9]*)?([eE][+-]?[0-9+])?</code>. On top of that, <code>NaN</code> and <code>Infinity</code> are also float constants.</li><li>A boolean constant is either <code>true</code> or <code>false</code> up to case-insensitivity (i.e. <code>True</code> is a valid boolean constant).</li><li>A <a href="http://en.wikipedia.org/wiki/Universally_unique_identifier">UUID</a> constant is defined b
y <code>hex{8}-hex{4}-hex{4}-hex{4}-hex{12}</code> where <code>hex</code> is an hexadecimal character, e.g. <code>[0-9a-fA-F]</code> and <code>{4}</code> is the number of such characters.</li><li>A blob constant is an hexadecimal number defined by <code>0[xX](hex)+</code> where <code>hex</code> is an hexadecimal character, e.g. <code>[0-9a-fA-F]</code>.</li></ul><p>For how these constants are typed, see the <a href="#types">data types section</a>.</p><h3 id="Comments">Comments</h3><p>A comment in CQL is a line beginning by either double dashes (<code>--</code>) or double slash (<code>//</code>).</p><p>Multi-line comments are also supported through enclosure within <code>/*</code> and <code>*/</code> (but nesting is not supported).</p><pre class="sample"><pre>-- This is a comment
// This is a comment too
@@ -101,7 +101,7 @@ CREATE TABLE timeline (
INSERT INTO test(pk, t, v, s) VALUES (0, 0, 'val0', 'static0');
INSERT INTO test(pk, t, v, s) VALUES (0, 1, 'val1', 'static1');
SELECT * FROM test WHERE pk=0 AND t=0;
-</pre></pre><p>the last query will return <code>'static1'</code> as value for <code>s</code>, since <code>s</code> is static and thus the 2nd insertion modified this “shared” value. Note however that static columns are only static within a given partition, and if in the example above both rows where from different partitions (i.e. if they had different value for <code>pk</code>), then the 2nd insertion would not have modified the value of <code>s</code> for the first row.</p><p>A few restrictions applies to when static columns are allowed:</p><ul><li>tables with the <code>COMPACT STORAGE</code> option (see below) cannot have them</li><li>a table without clustering columns cannot have static columns (in a table without clustering columns, every partition has only one row, and so every column is inherently static).</li><li>only non <code>PRIMARY KEY</code> columns can be static</li></ul><h4 id="createTableOptions"><code><option></code></h4><p>The <code>CREATE TABLE</cod
e> statement supports a number of options that controls the configuration of a new table. These options can be specified after the <code>WITH</code> keyword.</p><p>The first of these option is <code>COMPACT STORAGE</code>. This option is mainly targeted towards backward compatibility for definitions created before CQL3 (see <a href="http://www.datastax.com/dev/blog/thrift-to-cql3">www.datastax.com/dev/blog/thrift-to-cql3</a> for more details). The option also provides a slightly more compact layout of data on disk but at the price of diminished flexibility and extensibility for the table. Most notably, <code>COMPACT STORAGE</code> tables cannot have collections nor static columns and a <code>COMPACT STORAGE</code> table with at least one clustering column supports exactly one (as in not 0 nor more than 1) column not part of the <code>PRIMARY KEY</code> definition (which imply in particular that you cannot add nor remove columns after creation). For those reasons, <code>COMPACT STO
RAGE</code> is not recommended outside of the backward compatibility reason evoked above.</p><p>Another option is <code>CLUSTERING ORDER</code>. It allows to define the ordering of rows on disk. It takes the list of the clustering column names with, for each of them, the on-disk order (Ascending or descending). Note that this option affects <a href="#selectOrderBy">what <code>ORDER BY</code> are allowed during <code>SELECT</code></a>.</p><p>Table creation supports the following other <code><property></code>:</p><table><tr><th>option </th><th>kind </th><th>default </th><th>description</th></tr><tr><td><code>comment</code> </td><td><em>simple</em> </td><td>none </td><td>A free-form, human-readable comment.</td></tr><tr><td><code>read_repair_chance</code> </td><td><em>simple</em> </td><td>0.1 </td><td>The probability with which to query extra nodes (e.g. more nodes than required by the consistency level) for the purpos
e of read repairs.</td></tr><tr><td><code>dclocal_read_repair_chance</code> </td><td><em>simple</em> </td><td>0 </td><td>The probability with which to query extra nodes (e.g. more nodes than required by the consistency level) belonging to the same data center than the read coordinator for the purpose of read repairs.</td></tr><tr><td><code>gc_grace_seconds</code> </td><td><em>simple</em> </td><td>864000 </td><td>Time to wait before garbage collecting tombstones (deletion markers).</td></tr><tr><td><code>bloom_filter_fp_chance</code> </td><td><em>simple</em> </td><td>0.00075 </td><td>The target probability of false positive of the sstable bloom filters. Said bloom filters will be sized to provide the provided probability (thus lowering this value impact the size of bloom filters in-memory and on-disk)</td></tr><tr><td><code>compaction</code> </td><td><em>map</em> </td><td><em>see below</em> </td><td>The compaction options to use, se
e below.</td></tr><tr><td><code>compression</code> </td><td><em>map</em> </td><td><em>see below</em> </td><td>Compression options, see below. </td></tr><tr><td><code>replicate_on_write</code> </td><td><em>simple</em> </td><td>true </td><td>Whether to replicate data on write. This can only be set to false for tables with counters values. Disabling this is dangerous and can result in random lose of counters, don’t disable unless you are sure to know what you are doing</td></tr><tr><td><code>caching</code> </td><td><em>simple</em> </td><td>keys_only </td><td>Whether to cache keys (“key cache”) and/or rows (“row cache”) for this table. Valid values are: <code>all</code>, <code>keys_only</code>, <code>rows_only</code> and <code>none</code>. </td></tr></table><h4 id="compactionOptions"><code>compaction</code> options</h4><p>The <code>compaction</code> property must at least define the <code>'class'</code>
sub-option, that defines the compaction strategy class to use. The default supported class are <code>'SizeTieredCompactionStrategy'</code> and <code>'LeveledCompactionStrategy'</code>. Custom strategy can be provided by specifying the full class name as a <a href="#constants">string constant</a>. The rest of the sub-options depends on the chosen class. The sub-options supported by the default classes are:</p><table><tr><th>option </th><th>supported compaction strategy </th><th>default </th><th>description </th></tr><tr><td><code>tombstone_threshold</code> </td><td><em>all</em> </td><td>0.2 </td><td>A ratio such that if a sstable has more than this ratio of gcable tombstones over all contained columns, the sstable will be compacted (with no other sstables) for the purpose of purging those tombstones. </td></tr><tr><td><code>tombstone_compaction_interval</code> </td><td><em>all</em> </td><td>1 d
ay </td><td>The minimum time to wait after an sstable creation time before considering it for “tombstone compaction”, where “tombstone compaction” is the compaction triggered if the sstable has more gcable tombstones than <code>tombstone_threshold</code>. </td></tr><tr><td><code>min_sstable_size</code> </td><td>SizeTieredCompactionStrategy </td><td>50MB </td><td>The size tiered strategy groups SSTables to compact in buckets. A bucket groups SSTables that differs from less than 50% in size. However, for small sizes, this would result in a bucketing that is too fine grained. <code>min_sstable_size</code> defines a size threshold (in bytes) below which all SSTables belong to one unique bucket</td></tr><tr><td><code>min_threshold</code> </td><td>SizeTieredCompactionStrategy </td><td>4 </td><td>Minimum number of SSTables needed to start a minor compaction.</td></tr><tr><td><code>max_threshold</code>
</td><td>SizeTieredCompactionStrategy </td><td>32 </td><td>Maximum number of SSTables processed by one minor compaction.</td></tr><tr><td><code>bucket_low</code> </td><td>SizeTieredCompactionStrategy </td><td>0.5 </td><td>Size tiered consider sstables to be within the same bucket if their size is within [average_size * <code>bucket_low</code>, average_size * <code>bucket_high</code> ] (i.e the default groups sstable whose sizes diverges by at most 50%)</td></tr><tr><td><code>bucket_high</code> </td><td>SizeTieredCompactionStrategy </td><td>1.5 </td><td>Size tiered consider sstables to be within the same bucket if their size is within [average_size * <code>bucket_low</code>, average_size * <code>bucket_high</code> ] (i.e the default groups sstable whose sizes diverges by at most 50%).</td></tr><tr><td><code>sstable_size_in_mb</code> </td><td>LeveledCompactionStrategy </td><td>5MB </td><td>The
target size (in MB) for sstables in the leveled strategy. Note that while sstable sizes should stay less or equal to <code>sstable_size_in_mb</code>, it is possible to exceptionally have a larger sstable as during compaction, data for a given partition key are never split into 2 sstables</td></tr></table><p>For the <code>compression</code> property, the following default sub-options are available:</p><table><tr><th>option </th><th>default </th><th>description </th></tr><tr><td><code>sstable_compression</code> </td><td>LZ4Compressor </td><td>The compression algorithm to use. Default compressor are: LZ4Compressor, SnappyCompressor and DeflateCompressor. Use an empty string (<code>''</code>) to disable compression. Custom compressor can be provided by specifying the full class name as a <a href="#constants">string constant</a>.</td></tr><tr><td><code>chunk_length_kb</code> </td><td>64KB </td><td>On disk SSTables are compressed by block (to allow r
andom reads). This defines the size (in KB) of said block. Bigger values may improve the compression rate, but increases the minimum size of data to be read from disk for a read </td></tr><tr><td><code>crc_check_chance</code> </td><td>1.0 </td><td>When compression is enabled, each compressed block includes a checksum of that block for the purpose of detecting disk bitrot and avoiding the propagation of corruption to other replica. This option defines the probability with which those checksums are checked during read. By default they are always checked. Set to 0 to disable checksum checking and to 0.5 for instance to check them every other read</td></tr></table><h4 id="Otherconsiderations">Other considerations:</h4><ul><li>When <a href="#insertStmt/"updating":#updateStmt">inserting</a> a given row, not all columns needs to be defined (except for those part of the key), and missing columns occupy no space on disk. Furthermore, adding new columns (see <a hr
ef=#alterStmt><tt>ALTER TABLE</tt></a>) is a constant time operation. There is thus no need to try to anticipate future usage (or to cry when you haven’t) when creating a table.</li></ul><h3 id="alterTableStmt">ALTER TABLE</h3><p><i>Syntax:</i></p><pre class="syntax"><pre><alter-table-stmt> ::= ALTER (TABLE | COLUMNFAMILY) <tablename> <instruction>
+</pre></pre><p>the last query will return <code>'static1'</code> as value for <code>s</code>, since <code>s</code> is static and thus the 2nd insertion modified this “shared” value. Note however that static columns are only static within a given partition, and if in the example above both rows where from different partitions (i.e. if they had different value for <code>pk</code>), then the 2nd insertion would not have modified the value of <code>s</code> for the first row.</p><p>A few restrictions applies to when static columns are allowed:</p><ul><li>tables with the <code>COMPACT STORAGE</code> option (see below) cannot have them</li><li>a table without clustering columns cannot have static columns (in a table without clustering columns, every partition has only one row, and so every column is inherently static).</li><li>only non <code>PRIMARY KEY</code> columns can be static</li></ul><h4 id="createTableOptions"><code><option></code></h4><p>The <code>CREATE TABLE</cod
e> statement supports a number of options that controls the configuration of a new table. These options can be specified after the <code>WITH</code> keyword.</p><p>The first of these option is <code>COMPACT STORAGE</code>. This option is mainly targeted towards backward compatibility for definitions created before CQL3 (see <a href="http://www.datastax.com/dev/blog/thrift-to-cql3">www.datastax.com/dev/blog/thrift-to-cql3</a> for more details). The option also provides a slightly more compact layout of data on disk but at the price of diminished flexibility and extensibility for the table. Most notably, <code>COMPACT STORAGE</code> tables cannot have collections nor static columns and a <code>COMPACT STORAGE</code> table with at least one clustering column supports exactly one (as in not 0 nor more than 1) column not part of the <code>PRIMARY KEY</code> definition (which imply in particular that you cannot add nor remove columns after creation). For those reasons, <code>COMPACT STO
RAGE</code> is not recommended outside of the backward compatibility reason evoked above.</p><p>Another option is <code>CLUSTERING ORDER</code>. It allows to define the ordering of rows on disk. It takes the list of the clustering column names with, for each of them, the on-disk order (Ascending or descending). Note that this option affects <a href="#selectOrderBy">what <code>ORDER BY</code> are allowed during <code>SELECT</code></a>.</p><p>Table creation supports the following other <code><property></code>:</p><table><tr><th>option </th><th>kind </th><th>default </th><th>description</th></tr><tr><td><code>comment</code> </td><td><em>simple</em> </td><td>none </td><td>A free-form, human-readable comment.</td></tr><tr><td><code>read_repair_chance</code> </td><td><em>simple</em> </td><td>0.1 </td><td>The probability with which to query extra nodes (e.g. more nodes than required by the consistency level) for the purpos
e of read repairs.</td></tr><tr><td><code>dclocal_read_repair_chance</code> </td><td><em>simple</em> </td><td>0 </td><td>The probability with which to query extra nodes (e.g. more nodes than required by the consistency level) belonging to the same data center than the read coordinator for the purpose of read repairs.</td></tr><tr><td><code>gc_grace_seconds</code> </td><td><em>simple</em> </td><td>864000 </td><td>Time to wait before garbage collecting tombstones (deletion markers).</td></tr><tr><td><code>bloom_filter_fp_chance</code> </td><td><em>simple</em> </td><td>0.00075 </td><td>The target probability of false positive of the sstable bloom filters. Said bloom filters will be sized to provide the provided probability (thus lowering this value impact the size of bloom filters in-memory and on-disk)</td></tr><tr><td><code>compaction</code> </td><td><em>map</em> </td><td><em>see below</em> </td><td>The compaction options to use, se
e below.</td></tr><tr><td><code>compression</code> </td><td><em>map</em> </td><td><em>see below</em> </td><td>Compression options, see below. </td></tr><tr><td><code>replicate_on_write</code> </td><td><em>simple</em> </td><td>true </td><td>Whether to replicate data on write. This can only be set to false for tables with counters values. Disabling this is dangerous and can result in random lose of counters, don’t disable unless you are sure to know what you are doing</td></tr><tr><td><code>caching</code> </td><td><em>simple</em> </td><td>keys_only </td><td>Whether to cache keys (“key cache”) and/or rows (“row cache”) for this table. Valid values are: <code>all</code>, <code>keys_only</code>, <code>rows_only</code> and <code>none</code>. </td></tr></table><h4 id="compactionOptions"><code>compaction</code> options</h4><p>The <code>compaction</code> property must at least define the <code>'class'</code>
sub-option, that defines the compaction strategy class to use. The default supported class are <code>'SizeTieredCompactionStrategy'</code> and <code>'LeveledCompactionStrategy'</code>. Custom strategy can be provided by specifying the full class name as a <a href="#constants">string constant</a>. The rest of the sub-options depends on the chosen class. The sub-options supported by the default classes are:</p><table><tr><th>option </th><th>supported compaction strategy </th><th>default </th><th>description </th></tr><tr><td><code>enabled</code> </td><td><em>all</em> </td><td>true </td><td>A boolean denoting whether compaction should be enabled or not.</td></tr><tr><td><code>tombstone_threshold</code> </td><td><em>all</em> </td><td>0.2 </td><td>A ratio such that if a sstable has more than this ratio of gcable tombstones over all contained columns, the sstable will be c
ompacted (with no other sstables) for the purpose of purging those tombstones. </td></tr><tr><td><code>tombstone_compaction_interval</code> </td><td><em>all</em> </td><td>1 day </td><td>The minimum time to wait after an sstable creation time before considering it for “tombstone compaction”, where “tombstone compaction” is the compaction triggered if the sstable has more gcable tombstones than <code>tombstone_threshold</code>. </td></tr><tr><td><code>min_sstable_size</code> </td><td>SizeTieredCompactionStrategy </td><td>50MB </td><td>The size tiered strategy groups SSTables to compact in buckets. A bucket groups SSTables that differs from less than 50% in size. However, for small sizes, this would result in a bucketing that is too fine grained. <code>min_sstable_size</code> defines a size threshold (in bytes) below which all SSTables belong to one unique bucket</td></tr><tr><td><code>min_threshold</code>
</td><td>SizeTieredCompactionStrategy </td><td>4 </td><td>Minimum number of SSTables needed to start a minor compaction.</td></tr><tr><td><code>max_threshold</code> </td><td>SizeTieredCompactionStrategy </td><td>32 </td><td>Maximum number of SSTables processed by one minor compaction.</td></tr><tr><td><code>bucket_low</code> </td><td>SizeTieredCompactionStrategy </td><td>0.5 </td><td>Size tiered consider sstables to be within the same bucket if their size is within [average_size * <code>bucket_low</code>, average_size * <code>bucket_high</code> ] (i.e the default groups sstable whose sizes diverges by at most 50%)</td></tr><tr><td><code>bucket_high</code> </td><td>SizeTieredCompactionStrategy </td><td>1.5 </td><td>Size tiered consider sstables to be within the same bucket if their size is within [average_size * <code>bucket_low</code>, average_size * <code>bucket_high</code> ] (i.
e the default groups sstable whose sizes diverges by at most 50%).</td></tr><tr><td><code>sstable_size_in_mb</code> </td><td>LeveledCompactionStrategy </td><td>5MB </td><td>The target size (in MB) for sstables in the leveled strategy. Note that while sstable sizes should stay less or equal to <code>sstable_size_in_mb</code>, it is possible to exceptionally have a larger sstable as during compaction, data for a given partition key are never split into 2 sstables</td></tr></table><p>For the <code>compression</code> property, the following default sub-options are available:</p><table><tr><th>option </th><th>default </th><th>description </th></tr><tr><td><code>sstable_compression</code> </td><td>LZ4Compressor </td><td>The compression algorithm to use. Default compressor are: LZ4Compressor, SnappyCompressor and DeflateCompressor. Use an empty string (<code>''</code>) to disable compression. Custom compressor can be provided by specifying the
full class name as a <a href="#constants">string constant</a>.</td></tr><tr><td><code>chunk_length_kb</code> </td><td>64KB </td><td>On disk SSTables are compressed by block (to allow random reads). This defines the size (in KB) of said block. Bigger values may improve the compression rate, but increases the minimum size of data to be read from disk for a read </td></tr><tr><td><code>crc_check_chance</code> </td><td>1.0 </td><td>When compression is enabled, each compressed block includes a checksum of that block for the purpose of detecting disk bitrot and avoiding the propagation of corruption to other replica. This option defines the probability with which those checksums are checked during read. By default they are always checked. Set to 0 to disable checksum checking and to 0.5 for instance to check them every other read</td></tr></table><h4 id="Otherconsiderations">Other considerations:</h4><ul><li>When <a href="#insertStmt/"updating":#u
pdateStmt">inserting</a> a given row, not all columns needs to be defined (except for those part of the key), and missing columns occupy no space on disk. Furthermore, adding new columns (see <a href=#alterStmt><tt>ALTER TABLE</tt></a>) is a constant time operation. There is thus no need to try to anticipate future usage (or to cry when you haven’t) when creating a table.</li></ul><h3 id="alterTableStmt">ALTER TABLE</h3><p><i>Syntax:</i></p><pre class="syntax"><pre><alter-table-stmt> ::= ALTER (TABLE | COLUMNFAMILY) <tablename> <instruction>
<instruction> ::= ALTER <identifier> TYPE <type>
| ADD <identifier> <type>
@@ -146,13 +146,16 @@ USING TTL 86400;
( USING <option> ( AND <option> )* )?
SET <assignment> ( ',' <assignment> )*
WHERE <where-clause>
- ( IF <identifier> '=' <term> ( AND <identifier> '=' <term> )* )?
+ ( IF <condition> ( AND condition )* )?
<assignment> ::= <identifier> '=' <term>
| <identifier> '=' <identifier> ('+' | '-') (<int-term> | <set-literal> | <list-literal>)
| <identifier> '=' <identifier> '+' <map-literal>
| <identifier> '[' <term> ']' '=' <term>
+<condition> ::= <identifier> '=' <term>
+ | <identifier> '[' <term> ']' '=' <term>
+
<where-clause> ::= <relation> ( AND <relation> )*
<relation> ::= <identifier> '=' <term>
@@ -172,6 +175,7 @@ UPDATE UserActions SET total = total + 2
FROM <tablename>
( USING TIMESTAMP <integer>)?
WHERE <where-clause>
+ ( IF ( EXISTS | ( <condition> ( AND <condition> )*) ) )?
<selection> ::= <identifier> ( '[' <term> ']' )?
@@ -180,10 +184,13 @@ UPDATE UserActions SET total = total + 2
<relation> ::= <identifier> '=' <term>
| <identifier> IN '(' ( <term> ( ',' <term> )* )? ')'
| <identifier> IN '?'
+
+<condition> ::= <identifier> '=' <term>
+ | <identifier> '[' <term> ']' '=' <term>
</pre></pre><p><br/><i>Sample:</i></p><pre class="sample"><pre>DELETE FROM NerdMovies USING TIMESTAMP 1240003134 WHERE movie = 'Serenity';
DELETE phone FROM Users WHERE userid IN (C73DE1D3-AF08-40F3-B124-3FF3E5109F22, B70DE1D0-9908-4AE3-BE34-5573E5B09F14);
-</pre></pre><p><br/>The <code>DELETE</code> statement deletes columns and rows. If column names are provided directly after the <code>DELETE</code> keyword, only those columns are deleted from the row indicated by the <code><where-clause></code> (the <code>id[value]</code> syntax in <code><selection></code> is for collection, please refer to the <a href="#collections">collection section</a> for more details). Otherwise whole rows are removed. The <code><where-clause></code> allows to specify the key for the row(s) to delete (the <code>IN</code> relation is only supported for the last column of the partition key).</p><p><code>DELETE</code> supports the <code>TIMESTAMP</code> options with the same semantic that in the <a href="#updateStmt"><code>UPDATE</code></a> statement.</p><p>In a <code>DELETE</code> statement, all deletions within the same partition key are applied atomically and in isolation.</p><h3 id="batchStmt">BATCH</h3><p><i>Syntax:</i></p><pre class="syntax"><pre
><batch-stmt> ::= BEGIN ( UNLOGGED | COUNTER ) BATCH
+</pre></pre><p><br/>The <code>DELETE</code> statement deletes columns and rows. If column names are provided directly after the <code>DELETE</code> keyword, only those columns are deleted from the row indicated by the <code><where-clause></code> (the <code>id[value]</code> syntax in <code><selection></code> is for collection, please refer to the <a href="#collections">collection section</a> for more details). Otherwise whole rows are removed. The <code><where-clause></code> allows to specify the key for the row(s) to delete (the <code>IN</code> relation is only supported for the last column of the partition key).</p><p><code>DELETE</code> supports the <code>TIMESTAMP</code> options with the same semantic that in the <a href="#updateStmt"><code>UPDATE</code></a> statement.</p><p>In a <code>DELETE</code> statement, all deletions within the same partition key are applied atomically and in isolation.</p><p>A <code>DELETE</code> operation application can be conditioned using <c
ode>IF</code> like for <code>UPDATE</code> and <code>INSERT</code>. But please not that as for the later, this will incur a non negligible performance cost (internally, Paxos will be used) and so should be used sparingly.</p><h3 id="batchStmt">BATCH</h3><p><i>Syntax:</i></p><pre class="syntax"><pre><batch-stmt> ::= BEGIN ( UNLOGGED | COUNTER ) BATCH
( USING <option> ( AND <option> )* )?
<modification-stmt> ( ';' <modification-stmt> )*
APPLY BATCH
@@ -199,7 +206,7 @@ DELETE phone FROM Users WHERE userid IN
INSERT INTO users (userid, password) VALUES ('user4', 'ch@ngem3c');
DELETE name FROM users WHERE userid = 'user1';
APPLY BATCH;
-</pre></pre><p>The <code>BATCH</code> statement group multiple modification statements (insertions/updates and deletions) into a single statement. It serves several purposes:</p><ol><li>It saves network round-trips between the client and the server (and sometimes between the server coordinator and the replicas) when batching multiple updates.</li><li>All updates in a <code>BATCH</code> belonging to a given partition key are performed in isolation.</li><li>By default, all operations in the batch are performed atomically. See the notes on <a href="#unloggedBatch"><code>UNLOGGED</code></a> for more details.</li></ol><p>Note however that the <code>BATCH</code> statement only allows <code>UPDATE</code>, <code>INSERT</code> and <code>DELETE</code> statements and is <em>not</em> a full analogue for SQL transactions.</p><h4 id="unloggedBatch"><code>UNLOGGED</code></h4><p>By default, Cassandra uses a batch log to ensure all operations in a batch are applied atomically. (Note that the operat
ions are still only isolated within a single partition.)</p><p>There is a performance penalty for batch atomicity when a batch spans multiple partitions. If you do not want to incur this penalty, you can tell Cassandra to skip the batchlog with the <code>UNLOGGED</code> option. If the <code>UNLOGGED</code> option is used, operations are only atomic within a single partition.</p><h4 id="counterBatch"><code>COUNTER</code></h4><p>Use the <code>COUNTER</code> option for batched counter updates. Unlike other updates in Cassandra, counter updates are not idempotent.</p><h4 id="batchOptions"><code><option></code></h4><p><code>BATCH</code> supports both the <code>TIMESTAMP</code> option, with similar semantic to the one described in the <a href="#updateOptions"><code>UPDATE</code></a> statement (the timestamp applies to all the statement inside the batch). However, if used, <code>TIMESTAMP</code> <strong>must not</strong> be used in the statements within the batch.</p><h2 id="queries">Q
ueries</h2><h3 id="selectStmt">SELECT</h3><p><i>Syntax:</i></p><pre class="syntax"><pre><select-stmt> ::= SELECT <select-clause>
+</pre></pre><p>The <code>BATCH</code> statement group multiple modification statements (insertions/updates and deletions) into a single statement. It serves several purposes:</p><ol><li>It saves network round-trips between the client and the server (and sometimes between the server coordinator and the replicas) when batching multiple updates.</li><li>All updates in a <code>BATCH</code> belonging to a given partition key are performed in isolation.</li><li>By default, all operations in the batch are performed atomically. See the notes on <a href="#unloggedBatch"><code>UNLOGGED</code></a> for more details.</li></ol><p>Note that:</p><ul><li><code>BATCH</code> statements may only contain <code>UPDATE</code>, <code>INSERT</code> and <code>DELETE</code> statements.</li><li>Batches are <em>not</em> a full analogue for SQL transactions.</li><li>If a timestamp is not specified for each operation, then all operations will be applied with the same timestamp. Due to Cassandra’s conflict
resolution procedure in the case of <a href="http://wiki.apache.org/cassandra/FAQ#clocktie">timestamp ties</a>, operations may be applied in an order that is different from the order they are listed in the <code>BATCH</code> statement. To force a particular operation ordering, you must specify per-operation timestamps.</li></ul><h4 id="unloggedBatch"><code>UNLOGGED</code></h4><p>By default, Cassandra uses a batch log to ensure all operations in a batch are applied atomically. (Note that the operations are still only isolated within a single partition.)</p><p>There is a performance penalty for batch atomicity when a batch spans multiple partitions. If you do not want to incur this penalty, you can tell Cassandra to skip the batchlog with the <code>UNLOGGED</code> option. If the <code>UNLOGGED</code> option is used, operations are only atomic within a single partition.</p><h4 id="counterBatch"><code>COUNTER</code></h4><p>Use the <code>COUNTER</code> option for batched counter updates.
Unlike other updates in Cassandra, counter updates are not idempotent.</p><h4 id="batchOptions"><code><option></code></h4><p><code>BATCH</code> supports both the <code>TIMESTAMP</code> option, with similar semantic to the one described in the <a href="#updateOptions"><code>UPDATE</code></a> statement (the timestamp applies to all the statement inside the batch). However, if used, <code>TIMESTAMP</code> <strong>must not</strong> be used in the statements within the batch.</p><h2 id="queries">Queries</h2><h3 id="selectStmt">SELECT</h3><p><i>Syntax:</i></p><pre class="syntax"><pre><select-stmt> ::= SELECT <select-clause>
FROM <tablename>
( WHERE <where-clause> )?
( ORDER BY <order-by> )?
@@ -220,13 +227,15 @@ APPLY BATCH;
<where-clause> ::= <relation> ( AND <relation> )*
<relation> ::= <identifier> <op> <term>
- | '(' <identifier> (',' <identifier>)* ')' <op> '(' <term> (',' <term>)* ')'
+ | '(' <identifier> (',' <identifier>)* ')' <op> <term-tuple>
| <identifier> IN '(' ( <term> ( ',' <term>)* )? ')'
+ | '(' <identifier> (',' <identifier>)* ')' IN '(' ( <term-tuple> ( ',' <term-tuple>)* )? ')'
| TOKEN '(' <identifier> ( ',' <identifer>)* ')' <op> <term>
<op> ::= '=' | '<' | '>' | '<=' | '>='
<order-by> ::= <ordering> ( ',' <odering> )*
<ordering> ::= <identifer> ( ASC | DESC )?
+<term-tuple> ::= '(' <term> (',' <term>)* ')'
</pre></pre><p><br/><i>Sample:</i></p><pre class="sample"><pre>SELECT name, occupation FROM users WHERE userid IN (199, 200, 207);
SELECT name AS user_name, occupation AS user_occupation FROM users;
@@ -254,8 +263,9 @@ SELECT COUNT(*) AS user_count FROM users
</pre></pre><p>But the following one is not, as it does not select a contiguous set of rows (and we suppose no secondary indexes are set):</p><pre class="sample"><pre>// Needs a blog_title to be set to select ranges of posted_at
SELECT entry_title, content FROM posts WHERE userid='john doe' AND posted_at >= '2012-01-01' AND posted_at < '2012-01-31'
</pre></pre><p>When specifying relations, the <code>TOKEN</code> function can be used on the <code>PARTITION KEY</code> column to query. In that case, rows will be selected based on the token of their <code>PARTITION_KEY</code> rather than on the value. Note that the token of a key depends on the partitioner in use, and that in particular the RandomPartitioner won’t yield a meaningful order. Also note that ordering partitioners always order token values by bytes (so even if the partition key is of type int, <code>token(-1) > token(0)</code> in particular). Example:</p><pre class="sample"><pre>SELECT * FROM posts WHERE token(userid) > token('tom') AND token(userid) < token('bob')
-</pre></pre><p>Moreover, the <code>IN</code> relation is only allowed on the last column of the partition key and on the last column of the full primary key.</p><p>It is also possible to “group” <code>CLUSTERING COLUMNS</code> together in a relation, for instance:</p><pre class="sample"><pre>SELECT * FROM posts WHERE userid='john doe' AND (blog_title, posted_at) > ('John''s Blog', '2012-01-01')
+</pre></pre><p>Moreover, the <code>IN</code> relation is only allowed on the last column of the partition key and on the last column of the full primary key.</p><p>It is also possible to “group” <code>CLUSTERING COLUMNS</code> together in a relation using the tuple notation. For instance:</p><pre class="sample"><pre>SELECT * FROM posts WHERE userid='john doe' AND (blog_title, posted_at) > ('John''s Blog', '2012-01-01')
</pre></pre><p>will request all rows that sorts after the one having “John's Blog” as <code>blog_tile</code> and ‘2012-01-01’ for <code>posted_at</code> in the clustering order. In particular, rows having a <code>post_at <= '2012-01-01'</code> will be returned as long as their <code>blog_title > 'John''s Blog'</code>, which wouldn’t be the case for:</p><pre class="sample"><pre>SELECT * FROM posts WHERE userid='john doe' AND blog_title > 'John''s Blog' AND posted_at > '2012-01-01'
+</pre></pre><p>The tuple notation may also be used for <code>IN</code> clauses on <code>CLUSTERING COLUMNS</code>:</p><pre class="sample"><pre>SELECT * FROM posts WHERE userid='john doe' AND (blog_title, posted_at) IN (('John''s Blog', '2012-01-01), ('Extreme Chess', '2014-06-01'))
</pre></pre><h4 id="selectOrderBy"><code><order-by></code></h4><p>The <code>ORDER BY</code> option allows to select the order of the returned results. It takes as argument a list of column names along with the order for the column (<code>ASC</code> for ascendant and <code>DESC</code> for descendant, omitting the order being equivalent to <code>ASC</code>). Currently the possible orderings are limited (which depends on the table <a href="#createTableOptions"><code>CLUSTERING ORDER</code></a>):</p><ul><li>if the table has been defined without any specific <code>CLUSTERING ORDER</code>, then then allowed orderings are the order induced by the clustering columns and the reverse of that one.</li><li>otherwise, the orderings allowed are the order of the <code>CLUSTERING ORDER</code> option and the reversed one.</li></ul><h4 id="selectLimit"><code>LIMIT</code></h4><p>The <code>LIMIT</code> option to a <code>SELECT</code> statement limits the number of rows returned by a query.</p><h4 id
="selectAllowFiltering"><code>ALLOW FILTERING</code></h4><p>By default, CQL only allows select queries that don’t involve “filtering” server side, i.e. queries where we know that all (live) record read will be returned (maybe partly) in the result set. The reasoning is that those “non filtering” queries have predictable performance in the sense that they will execute in a time that is proportional to the amount of data <strong>returned</strong> by the query (which can be controlled through <code>LIMIT</code>).</p><p>The <code>ALLOW FILTERING</code> option allows to explicitly allow (some) queries that require filtering. Please note that a query using <code>ALLOW FILTERING</code> may thus have unpredictable performance (for the definition above), i.e. even a query that selects a handful of records <strong>may</strong> exhibit performance that depends on the total amount of data stored in the cluster.</p><p>For instance, considering the following table ho
lding user profiles with their year of birth (with a secondary index on it) and country of residence:</p><pre class="sample"><pre>CREATE TABLE users (
username text PRIMARY KEY,
firstname text,
@@ -338,4 +348,4 @@ UPDATE plays SET scores = scores - [ 12,
)
</pre></pre><p>then the <code>token</code> function will take a single argument of type <code>text</code> (in that case, the partition key is <code>userid</code> (there is no clustering columns so the partition key is the same than the primary key)), and the return type will be <code>bigint</code>.</p><h3 id="uuidFun">Uuid</h3><p>The <code>uuid</code> function takes no parameters and generates a random type 4 uuid suitable for use in INSERT or SET statements.</p><h3 id="timeuuidFun">Timeuuid functions</h3><h4 id="now"><code>now</code></h4><p>The <code>now</code> function takes no arguments and generates a new unique timeuuid (at the time where the statement using it is executed). Note that this method is useful for insertion but is largely non-sensical in <code>WHERE</code> clauses. For instance, a query of the form</p><pre class="sample"><pre>SELECT * FROM myTable WHERE t = now()
</pre></pre><p>will never return any result by design, since the value returned by <code>now()</code> is guaranteed to be unique.</p><h4 id="minTimeuuidandmaxTimeuuid"><code>minTimeuuid</code> and <code>maxTimeuuid</code></h4><p>The <code>minTimeuuid</code> (resp. <code>maxTimeuuid</code>) function takes a <code>timestamp</code> value <code>t</code> (which can be <a href="#usingdates">either a timestamp or a date string</a>) and return a <em>fake</em> <code>timeuuid</code> corresponding to the <em>smallest</em> (resp. <em>biggest</em>) possible <code>timeuuid</code> having for timestamp <code>t</code>. So for instance:</p> <pre class="sample"><pre>SELECT * FROM myTable WHERE t > maxTimeuuid('2013-01-01 00:05+0000') AND t < minTimeuuid('2013-02-02 10:00+0000')
-</pre></pre> <p>will select all rows where the <code>timeuuid</code> column <code>t</code> is strictly older than ‘2013-01-01 00:05+0000’ but strictly younger than ‘2013-02-02 10:00+0000’. Please note that <code>t >= maxTimeuuid('2013-01-01 00:05+0000')</code> would still <em>not</em> select a <code>timeuuid</code> generated exactly at ‘2013-01-01 00:05+0000’ and is essentially equivalent to <code>t > maxTimeuuid('2013-01-01 00:05+0000')</code>.</p><p><em>Warning</em>: We called the values generated by <code>minTimeuuid</code> and <code>maxTimeuuid</code> <em>fake</em> UUID because they do no respect the Time-Based UUID generation process specified by the <a href="http://www.ietf.org/rfc/rfc4122.txt">RFC 4122</a>. In particular, the value returned by these 2 methods will not be unique. This means you should only use those methods for querying (as in the example above). Inserting the result of those methods is almost certainly <em>a bad idea</em>.
</p><h4 id="dateOfandunixTimestampOf"><code>dateOf</code> and <code>unixTimestampOf</code></h4><p>The <code>dateOf</code> and <code>unixTimestampOf</code> functions take a <code>timeuuid</code> argument and extract the embedded timestamp. However, while the <code>dateof</code> function return it with the <code>timestamp</code> type (that most client, including cqlsh, interpret as a date), the <code>unixTimestampOf</code> function returns it as a <code>bigint</code> raw value.</p><h3 id="blobFun">Blob conversion functions</h3><p>A number of functions are provided to “convert” the native types into binary data (<code>blob</code>). For every <code><native-type></code> <code>type</code> supported by CQL3 (a notable exceptions is <code>blob</code>, for obvious reasons), the function <code>typeAsBlob</code> takes a argument of type <code>type</code> and return it as a <code>blob</code>. Conversely, the function <code>blobAsType</code> takes a 64-bit <code>blob</code> argum
ent and convert it to a <code>bigint</code> value. And so for instance, <code>bigintAsBlob(3)</code> is <code>0x0000000000000003</code> and <code>blobAsBigint(0x0000000000000003)</code> is <code>3</code>.</p><h2 id="appendixA">Appendix A: CQL Keywords</h2><p>CQL distinguishes between <em>reserved</em> and <em>non-reserved</em> keywords. Reserved keywords cannot be used as identifier, they are truly reserved for the language (but one can enclose a reserved keyword by double-quotes to use it as an identifier). Non-reserved keywords however only have a specific meaning in certain context but can used as identifer otherwise. The only <em>raison d'être</em> of these non-reserved keywords is convenience: some keyword are non-reserved when it was always easy for the parser to decide whether they were used as keywords or not.</p><table><tr><th>Keyword </th><th>Reserved? </th></tr><tr><td><code>ADD</code> </td><td>yes </td></tr><tr><td><code>ALL</code> </td><td>no
</td></tr><tr><td><code>ALTER</code> </td><td>yes </td></tr><tr><td><code>AND</code> </td><td>yes </td></tr><tr><td><code>ANY</code> </td><td>yes </td></tr><tr><td><code>APPLY</code> </td><td>yes </td></tr><tr><td><code>AS</code> </td><td>no </td></tr><tr><td><code>ASC</code> </td><td>yes </td></tr><tr><td><code>ASCII</code> </td><td>no </td></tr><tr><td><code>AUTHORIZE</code> </td><td>yes </td></tr><tr><td><code>BATCH</code> </td><td>yes </td></tr><tr><td><code>BEGIN</code> </td><td>yes </td></tr><tr><td><code>BIGINT</code> </td><td>no </td></tr><tr><td><code>BLOB</code> </td><td>no </td></tr><tr><td><code>BOOLEAN</code> </td><td>no </td></tr><tr><td><code>BY</code> </td><td>yes </td></tr><tr><td><code>CLUSTERING</code> </td><td>no </td></tr><tr><td><code>COLUMNFAMILY</code> </td><td>yes </td></tr><tr><td><code>COMPACT</code> </td><td>no </td></tr><tr><td><code>CON
SISTENCY</code> </td><td>no </td></tr><tr><td><code>COUNT</code> </td><td>no </td></tr><tr><td><code>COUNTER</code> </td><td>no </td></tr><tr><td><code>CREATE</code> </td><td>yes </td></tr><tr><td><code>DECIMAL</code> </td><td>no </td></tr><tr><td><code>DELETE</code> </td><td>yes </td></tr><tr><td><code>DESC</code> </td><td>yes </td></tr><tr><td><code>DOUBLE</code> </td><td>no </td></tr><tr><td><code>DROP</code> </td><td>yes </td></tr><tr><td><code>EACH_QUORUM</code> </td><td>yes </td></tr><tr><td><code>FLOAT</code> </td><td>no </td></tr><tr><td><code>FROM</code> </td><td>yes </td></tr><tr><td><code>GRANT</code> </td><td>yes </td></tr><tr><td><code>IN</code> </td><td>yes </td></tr><tr><td><code>INDEX</code> </td><td>yes </td></tr><tr><td><code>CUSTOM</code> </td><td>no </td></tr><tr><td><code>INSERT</code> </td><td>yes </td></tr><tr><td><code>INT</code> </td><td>no
</td></tr><tr><td><code>INTO</code> </td><td>yes </td></tr><tr><td><code>KEY</code> </td><td>no </td></tr><tr><td><code>KEYSPACE</code> </td><td>yes </td></tr><tr><td><code>LEVEL</code> </td><td>no </td></tr><tr><td><code>LIMIT</code> </td><td>yes </td></tr><tr><td><code>LOCAL_ONE</code> </td><td>yes </td></tr><tr><td><code>LOCAL_QUORUM</code> </td><td>yes </td></tr><tr><td><code>MODIFY</code> </td><td>yes </td></tr><tr><td><code>NORECURSIVE</code> </td><td>yes </td></tr><tr><td><code>NOSUPERUSER</code> </td><td>no </td></tr><tr><td><code>OF</code> </td><td>yes </td></tr><tr><td><code>ON</code> </td><td>yes </td></tr><tr><td><code>ONE</code> </td><td>yes </td></tr><tr><td><code>ORDER</code> </td><td>yes </td></tr><tr><td><code>PASSWORD</code> </td><td>no </td></tr><tr><td><code>PERMISSION</code> </td><td>no </td></tr><tr><td><code>PERMISSIONS</code> </td><td>no </td></tr><tr><td><code>PR
IMARY</code> </td><td>yes </td></tr><tr><td><code>QUORUM</code> </td><td>yes </td></tr><tr><td><code>REVOKE</code> </td><td>yes </td></tr><tr><td><code>SCHEMA</code> </td><td>yes </td></tr><tr><td><code>SELECT</code> </td><td>yes </td></tr><tr><td><code>SET</code> </td><td>yes </td></tr><tr><td><code>STORAGE</code> </td><td>no </td></tr><tr><td><code>SUPERUSER</code> </td><td>no </td></tr><tr><td><code>TABLE</code> </td><td>yes </td></tr><tr><td><code>TEXT</code> </td><td>no </td></tr><tr><td><code>TIMESTAMP</code> </td><td>no </td></tr><tr><td><code>TIMEUUID</code> </td><td>no </td></tr><tr><td><code>THREE</code> </td><td>yes </td></tr><tr><td><code>TOKEN</code> </td><td>yes </td></tr><tr><td><code>TRUNCATE</code> </td><td>yes </td></tr><tr><td><code>TTL</code> </td><td>no </td></tr><tr><td><code>TWO</code> </td><td>yes </td></tr><tr><td><code>TYPE</code> </td><td>n
o </td></tr><tr><td><code>UPDATE</code> </td><td>yes </td></tr><tr><td><code>USE</code> </td><td>yes </td></tr><tr><td><code>USER</code> </td><td>no </td></tr><tr><td><code>USERS</code> </td><td>no </td></tr><tr><td><code>USING</code> </td><td>yes </td></tr><tr><td><code>UUID</code> </td><td>no </td></tr><tr><td><code>VALUES</code> </td><td>no </td></tr><tr><td><code>VARCHAR</code> </td><td>no </td></tr><tr><td><code>VARINT</code> </td><td>no </td></tr><tr><td><code>WHERE</code> </td><td>yes </td></tr><tr><td><code>WITH</code> </td><td>yes </td></tr><tr><td><code>WRITETIME</code> </td><td>no </td></tr><tr><td><code>DISTINCT</code> </td><td>no </td></tr></table><h2 id="changes">Changes</h2><p>The following describes the addition/changes brought for each version of CQL.</p><h3 id="a3.1.5">3.1.5</h3><ul><li>It is now possible to group clustering columns in a relatiion, see <a href="#selectWhere
">SELECT Where clauses</a>.</li><li>Added support for <code>STATIC</code> columns, see <a href="#createTableStatic">static in CREATE TABLE</a>.</li></ul><h3 id="a3.1.4">3.1.4</h3><ul><li><code>CREATE INDEX</code> now allows specifying options when creating CUSTOM indexes (see <a href="#createIndexStmt">CREATE INDEX reference</a>).</li></ul><h3 id="a3.1.3">3.1.3</h3><ul><li>Millisecond precision formats have been added to the timestamp parser (see <a href="#usingdates">working with dates</a>).</li></ul><h3 id="a3.1.2">3.1.2</h3><ul><li><code>NaN</code> and <code>Infinity</code> has been added as valid float contants. They are now reserved keywords. In the unlikely case you we using them as a column identifier (or keyspace/table one), you will noew need to double quote them (see <a href="#identifiers">quote identifiers</a>).</li></ul><h3 id="a3.1.1">3.1.1</h3><ul><li><code>SELECT</code> statement now allows listing the partition keys (using the <code>DISTINCT</code> modifier). See <a
href="https://issues.apache.org/jira/browse/CASSANDRA-4536">CASSANDRA-4536</a>.</li><li>The syntax <code>c IN ?</code> is now supported in <code>WHERE</code> clauses. In that case, the value expected for the bind variable will be a list of whatever type <code>c</code> is.</li><li>It is now possible to use named bind variables (using <code>:name</code> instead of <code>?</code>).</li></ul><h3 id="a3.1.0">3.1.0</h3><ul><li><a href="#alterTableStmt">ALTER TABLE</a> <code>DROP</code> option has been reenabled for CQL3 tables and has new semantics now: the space formerly used by dropped columns will now be eventually reclaimed (post-compaction). You should not readd previously dropped columns unless you use timestamps with microsecond precision (see <a href="https://issues.apache.org/jira/browse/CASSANDRA-3919">CASSANDRA-3919</a> for more details).</li><li><code>SELECT</code> statement now supports aliases in select clause. Aliases in WHERE and ORDER BY clauses are not supported. See the
“section on select”#selectStmt for details.</li><li><code>CREATE</code> statements for <code>KEYSPACE</code>, <code>TABLE</code> and <code>INDEX</code> now supports an <code>IF NOT EXISTS</code> condition. Similarly, <code>DROP</code> statements support a <code>IF EXISTS</code> condition.</li><li><code>INSERT</code> statements optionally supports a <code>IF NOT EXISTS</code> condition and <code>UPDATE</code> supports <code>IF</code> conditions.</li></ul><h3 id="a3.0.5">3.0.5</h3><ul><li><code>SELECT</code>, <code>UPDATE</code>, and <code>DELETE</code> statements now allow empty <code>IN</code> relations (see <a href="https://issues.apache.org/jira/browse/CASSANDRA-5626">CASSANDRA-5626</a>).</li></ul><h3 id="a3.0.4">3.0.4</h3><ul><li>Updated the syntax for custom <a href="#createIndexStmt">secondary indexes</a>.</li><li>Non-equal condition on the partition key are now never supported, even for ordering partitioner as this was not correct (the order was <strong>not</stron
g> the one of the type of the partition key). Instead, the <code>token</code> method should always be used for range queries on the partition key (see <a href="#selectWhere">WHERE clauses</a>).</li></ul><h3 id="a3.0.3">3.0.3</h3><ul><li>Support for custom <a href="#createIndexStmt">secondary indexes</a> has been added.</li></ul><h3 id="a3.0.2">3.0.2</h3><ul><li>Type validation for the <a href="#constants">constants</a> has been fixed. For instance, the implementation used to allow <code>'2'</code> as a valid value for an <code>int</code> column (interpreting it has the equivalent of <code>2</code>), or <code>42</code> as a valid <code>blob</code> value (in which case <code>42</code> was interpreted as an hexadecimal representation of the blob). This is no longer the case, type validation of constants is now more strict. See the <a href="#types">data types</a> section for details on which constant is allowed for which type.</li><li>The type validation fixed of the previous point has
lead to the introduction of <a href="#constants">blobs constants</a> to allow inputing blobs. Do note that while inputing blobs as strings constant is still supported by this version (to allow smoother transition to blob constant), it is now deprecated (in particular the <a href="#types">data types</a> section does not list strings constants as valid blobs) and will be removed by a future version. If you were using strings as blobs, you should thus update your client code ASAP to switch blob constants.</li><li>A number of functions to convert native types to blobs have also been introduced. Furthermore the token function is now also allowed in select clauses. See the <a href="#functions">section on functions</a> for details.</li></ul><h3 id="a3.0.1">3.0.1</h3><ul><li><a href="#usingdates">Date strings</a> (and timestamps) are no longer accepted as valid <code>timeuuid</code> values. Doing so was a bug in the sense that date string are not valid <code>timeuuid</code>, and it was thus
resulting in <a href="https://issues.apache.org/jira/browse/CASSANDRA-4936">confusing behaviors</a>. However, the following new methods have been added to help working with <code>timeuuid</code>: <code>now</code>, <code>minTimeuuid</code>, <code>maxTimeuuid</code> , <code>dateOf</code> and <code>unixTimestampOf</code>. See the <a href="#usingtimeuuid">section dedicated to these methods</a> for more detail.</li><li>“Float constants”#constants now support the exponent notation. In other words, <code>4.2E10</code> is now a valid floating point value.</li></ul><h2 id="Versioning">Versioning</h2><p>Versioning of the CQL language adheres to the <a href="http://semver.org">Semantic Versioning</a> guidelines. Versions take the form X.Y.Z where X, Y, and Z are integer values representing major, minor, and patch level respectively. There is no correlation between Cassandra release versions and the CQL language version.</p><table><tr><th>version</th><th>description</th></tr><tr><
td>Major </td><td>The major version <em>must</em> be bumped when backward incompatible changes are introduced. This should rarely occur.</td></tr><tr><td>Minor </td><td>Minor version increments occur when new, but backward compatible, functionality is introduced.</td></tr><tr><td>Patch </td><td>The patch version is incremented when bugs are fixed.</td></tr></table></body></html>
\ No newline at end of file
+</pre></pre> <p>will select all rows where the <code>timeuuid</code> column <code>t</code> is strictly older than ‘2013-01-01 00:05+0000’ but strictly younger than ‘2013-02-02 10:00+0000’. Please note that <code>t >= maxTimeuuid('2013-01-01 00:05+0000')</code> would still <em>not</em> select a <code>timeuuid</code> generated exactly at ‘2013-01-01 00:05+0000’ and is essentially equivalent to <code>t > maxTimeuuid('2013-01-01 00:05+0000')</code>.</p><p><em>Warning</em>: We called the values generated by <code>minTimeuuid</code> and <code>maxTimeuuid</code> <em>fake</em> UUID because they do no respect the Time-Based UUID generation process specified by the <a href="http://www.ietf.org/rfc/rfc4122.txt">RFC 4122</a>. In particular, the value returned by these 2 methods will not be unique. This means you should only use those methods for querying (as in the example above). Inserting the result of those methods is almost certainly <em>a bad idea</em>.
</p><h4 id="dateOfandunixTimestampOf"><code>dateOf</code> and <code>unixTimestampOf</code></h4><p>The <code>dateOf</code> and <code>unixTimestampOf</code> functions take a <code>timeuuid</code> argument and extract the embedded timestamp. However, while the <code>dateof</code> function return it with the <code>timestamp</code> type (that most client, including cqlsh, interpret as a date), the <code>unixTimestampOf</code> function returns it as a <code>bigint</code> raw value.</p><h3 id="blobFun">Blob conversion functions</h3><p>A number of functions are provided to “convert” the native types into binary data (<code>blob</code>). For every <code><native-type></code> <code>type</code> supported by CQL3 (a notable exceptions is <code>blob</code>, for obvious reasons), the function <code>typeAsBlob</code> takes a argument of type <code>type</code> and return it as a <code>blob</code>. Conversely, the function <code>blobAsType</code> takes a 64-bit <code>blob</code> argum
ent and convert it to a <code>bigint</code> value. And so for instance, <code>bigintAsBlob(3)</code> is <code>0x0000000000000003</code> and <code>blobAsBigint(0x0000000000000003)</code> is <code>3</code>.</p><h2 id="appendixA">Appendix A: CQL Keywords</h2><p>CQL distinguishes between <em>reserved</em> and <em>non-reserved</em> keywords. Reserved keywords cannot be used as identifier, they are truly reserved for the language (but one can enclose a reserved keyword by double-quotes to use it as an identifier). Non-reserved keywords however only have a specific meaning in certain context but can used as identifer otherwise. The only <em>raison d'être</em> of these non-reserved keywords is convenience: some keyword are non-reserved when it was always easy for the parser to decide whether they were used as keywords or not.</p><table><tr><th>Keyword </th><th>Reserved? </th></tr><tr><td><code>ADD</code> </td><td>yes </td></tr><tr><td><code>ALL</code> </td><td>no
</td></tr><tr><td><code>ALTER</code> </td><td>yes </td></tr><tr><td><code>AND</code> </td><td>yes </td></tr><tr><td><code>ANY</code> </td><td>yes </td></tr><tr><td><code>APPLY</code> </td><td>yes </td></tr><tr><td><code>AS</code> </td><td>no </td></tr><tr><td><code>ASC</code> </td><td>yes </td></tr><tr><td><code>ASCII</code> </td><td>no </td></tr><tr><td><code>AUTHORIZE</code> </td><td>yes </td></tr><tr><td><code>BATCH</code> </td><td>yes </td></tr><tr><td><code>BEGIN</code> </td><td>yes </td></tr><tr><td><code>BIGINT</code> </td><td>no </td></tr><tr><td><code>BLOB</code> </td><td>no </td></tr><tr><td><code>BOOLEAN</code> </td><td>no </td></tr><tr><td><code>BY</code> </td><td>yes </td></tr><tr><td><code>CLUSTERING</code> </td><td>no </td></tr><tr><td><code>COLUMNFAMILY</code> </td><td>yes </td></tr><tr><td><code>COMPACT</code> </td><td>no </td></tr><tr><td><code>CON
SISTENCY</code> </td><td>no </td></tr><tr><td><code>COUNT</code> </td><td>no </td></tr><tr><td><code>COUNTER</code> </td><td>no </td></tr><tr><td><code>CREATE</code> </td><td>yes </td></tr><tr><td><code>DECIMAL</code> </td><td>no </td></tr><tr><td><code>DELETE</code> </td><td>yes </td></tr><tr><td><code>DESC</code> </td><td>yes </td></tr><tr><td><code>DOUBLE</code> </td><td>no </td></tr><tr><td><code>DROP</code> </td><td>yes </td></tr><tr><td><code>EACH_QUORUM</code> </td><td>yes </td></tr><tr><td><code>FLOAT</code> </td><td>no </td></tr><tr><td><code>FROM</code> </td><td>yes </td></tr><tr><td><code>GRANT</code> </td><td>yes </td></tr><tr><td><code>IN</code> </td><td>yes </td></tr><tr><td><code>INDEX</code> </td><td>yes </td></tr><tr><td><code>CUSTOM</code> </td><td>no </td></tr><tr><td><code>INSERT</code> </td><td>yes </td></tr><tr><td><code>INT</code> </td><td>no
</td></tr><tr><td><code>INTO</code> </td><td>yes </td></tr><tr><td><code>KEY</code> </td><td>no </td></tr><tr><td><code>KEYSPACE</code> </td><td>yes </td></tr><tr><td><code>LEVEL</code> </td><td>no </td></tr><tr><td><code>LIMIT</code> </td><td>yes </td></tr><tr><td><code>LOCAL_ONE</code> </td><td>yes </td></tr><tr><td><code>LOCAL_QUORUM</code> </td><td>yes </td></tr><tr><td><code>MODIFY</code> </td><td>yes </td></tr><tr><td><code>NORECURSIVE</code> </td><td>yes </td></tr><tr><td><code>NOSUPERUSER</code> </td><td>no </td></tr><tr><td><code>OF</code> </td><td>yes </td></tr><tr><td><code>ON</code> </td><td>yes </td></tr><tr><td><code>ONE</code> </td><td>yes </td></tr><tr><td><code>ORDER</code> </td><td>yes </td></tr><tr><td><code>PASSWORD</code> </td><td>no </td></tr><tr><td><code>PERMISSION</code> </td><td>no </td></tr><tr><td><code>PERMISSIONS</code> </td><td>no </td></tr><tr><td><code>PR
IMARY</code> </td><td>yes </td></tr><tr><td><code>QUORUM</code> </td><td>yes </td></tr><tr><td><code>REVOKE</code> </td><td>yes </td></tr><tr><td><code>SCHEMA</code> </td><td>yes </td></tr><tr><td><code>SELECT</code> </td><td>yes </td></tr><tr><td><code>SET</code> </td><td>yes </td></tr><tr><td><code>STORAGE</code> </td><td>no </td></tr><tr><td><code>SUPERUSER</code> </td><td>no </td></tr><tr><td><code>TABLE</code> </td><td>yes </td></tr><tr><td><code>TEXT</code> </td><td>no </td></tr><tr><td><code>TIMESTAMP</code> </td><td>no </td></tr><tr><td><code>TIMEUUID</code> </td><td>no </td></tr><tr><td><code>THREE</code> </td><td>yes </td></tr><tr><td><code>TOKEN</code> </td><td>yes </td></tr><tr><td><code>TRUNCATE</code> </td><td>yes </td></tr><tr><td><code>TTL</code> </td><td>no </td></tr><tr><td><code>TWO</code> </td><td>yes </td></tr><tr><td><code>TYPE</code> </td><td>n
o </td></tr><tr><td><code>UPDATE</code> </td><td>yes </td></tr><tr><td><code>USE</code> </td><td>yes </td></tr><tr><td><code>USER</code> </td><td>no </td></tr><tr><td><code>USERS</code> </td><td>no </td></tr><tr><td><code>USING</code> </td><td>yes </td></tr><tr><td><code>UUID</code> </td><td>no </td></tr><tr><td><code>VALUES</code> </td><td>no </td></tr><tr><td><code>VARCHAR</code> </td><td>no </td></tr><tr><td><code>VARINT</code> </td><td>no </td></tr><tr><td><code>WHERE</code> </td><td>yes </td></tr><tr><td><code>WITH</code> </td><td>yes </td></tr><tr><td><code>WRITETIME</code> </td><td>no </td></tr><tr><td><code>DISTINCT</code> </td><td>no </td></tr></table><h2 id="changes">Changes</h2><p>The following describes the addition/changes brought for each version of CQL.</p><h3 id="a3.1.7">3.1.7</h3><ul><li><code>SELECT</code> statements now support selecting multiple rows in a single partition
using an <code>IN</code> clause on combinations of clustering columns. See <a href="#selectWhere">SELECT WHERE</a> clauses.</li></ul><h3 id="a3.1.6">3.1.6</h3><ul><li>A new <a href="#uuidFun"><code>uuid</code> method</a> has been added.</li><li>Support for <code>DELETE ... IF EXISTS</code> syntax.</li></ul><h3 id="a3.1.5">3.1.5</h3><ul><li>It is now possible to group clustering columns in a relatiion, see <a href="#selectWhere">SELECT WHERE</a> clauses.</li><li>Added support for <code>STATIC</code> columns, see <a href="#createTableStatic">static in CREATE TABLE</a>.</li></ul><h3 id="a3.1.4">3.1.4</h3><ul><li><code>CREATE INDEX</code> now allows specifying options when creating CUSTOM indexes (see <a href="#createIndexStmt">CREATE INDEX reference</a>).</li></ul><h3 id="a3.1.3">3.1.3</h3><ul><li>Millisecond precision formats have been added to the timestamp parser (see <a href="#usingdates">working with dates</a>).</li></ul><h3 id="a3.1.2">3.1.2</h3><ul><li><code>NaN</code> and <co
de>Infinity</code> has been added as valid float contants. They are now reserved keywords. In the unlikely case you we using them as a column identifier (or keyspace/table one), you will noew need to double quote them (see <a href="#identifiers">quote identifiers</a>).</li></ul><h3 id="a3.1.1">3.1.1</h3><ul><li><code>SELECT</code> statement now allows listing the partition keys (using the <code>DISTINCT</code> modifier). See <a href="https://issues.apache.org/jira/browse/CASSANDRA-4536">CASSANDRA-4536</a>.</li><li>The syntax <code>c IN ?</code> is now supported in <code>WHERE</code> clauses. In that case, the value expected for the bind variable will be a list of whatever type <code>c</code> is.</li><li>It is now possible to use named bind variables (using <code>:name</code> instead of <code>?</code>).</li></ul><h3 id="a3.1.0">3.1.0</h3><ul><li><a href="#alterTableStmt">ALTER TABLE</a> <code>DROP</code> option has been reenabled for CQL3 tables and has new semantics now: the space f
ormerly used by dropped columns will now be eventually reclaimed (post-compaction). You should not readd previously dropped columns unless you use timestamps with microsecond precision (see <a href="https://issues.apache.org/jira/browse/CASSANDRA-3919">CASSANDRA-3919</a> for more details).</li><li><code>SELECT</code> statement now supports aliases in select clause. Aliases in WHERE and ORDER BY clauses are not supported. See the “section on select”#selectStmt for details.</li><li><code>CREATE</code> statements for <code>KEYSPACE</code>, <code>TABLE</code> and <code>INDEX</code> now supports an <code>IF NOT EXISTS</code> condition. Similarly, <code>DROP</code> statements support a <code>IF EXISTS</code> condition.</li><li><code>INSERT</code> statements optionally supports a <code>IF NOT EXISTS</code> condition and <code>UPDATE</code> supports <code>IF</code> conditions.</li></ul><h3 id="a3.0.5">3.0.5</h3><ul><li><code>SELECT</code>, <code>UPDATE</code>, and <code>DELETE</
code> statements now allow empty <code>IN</code> relations (see <a href="https://issues.apache.org/jira/browse/CASSANDRA-5626">CASSANDRA-5626</a>).</li></ul><h3 id="a3.0.4">3.0.4</h3><ul><li>Updated the syntax for custom <a href="#createIndexStmt">secondary indexes</a>.</li><li>Non-equal condition on the partition key are now never supported, even for ordering partitioner as this was not correct (the order was <strong>not</strong> the one of the type of the partition key). Instead, the <code>token</code> method should always be used for range queries on the partition key (see <a href="#selectWhere">WHERE clauses</a>).</li></ul><h3 id="a3.0.3">3.0.3</h3><ul><li>Support for custom <a href="#createIndexStmt">secondary indexes</a> has been added.</li></ul><h3 id="a3.0.2">3.0.2</h3><ul><li>Type validation for the <a href="#constants">constants</a> has been fixed. For instance, the implementation used to allow <code>'2'</code> as a valid value for an <code>int</code> column (interpreting
it has the equivalent of <code>2</code>), or <code>42</code> as a valid <code>blob</code> value (in which case <code>42</code> was interpreted as an hexadecimal representation of the blob). This is no longer the case, type validation of constants is now more strict. See the <a href="#types">data types</a> section for details on which constant is allowed for which type.</li><li>The type validation fixed of the previous point has lead to the introduction of <a href="#constants">blobs constants</a> to allow inputing blobs. Do note that while inputing blobs as strings constant is still supported by this version (to allow smoother transition to blob constant), it is now deprecated (in particular the <a href="#types">data types</a> section does not list strings constants as valid blobs) and will be removed by a future version. If you were using strings as blobs, you should thus update your client code ASAP to switch blob constants.</li><li>A number of functions to convert native types to
blobs have also been introduced. Furthermore the token function is now also allowed in select clauses. See the <a href="#functions">section on functions</a> for details.</li></ul><h3 id="a3.0.1">3.0.1</h3><ul><li><a href="#usingdates">Date strings</a> (and timestamps) are no longer accepted as valid <code>timeuuid</code> values. Doing so was a bug in the sense that date string are not valid <code>timeuuid</code>, and it was thus resulting in <a href="https://issues.apache.org/jira/browse/CASSANDRA-4936">confusing behaviors</a>. However, the following new methods have been added to help working with <code>timeuuid</code>: <code>now</code>, <code>minTimeuuid</code>, <code>maxTimeuuid</code> , <code>dateOf</code> and <code>unixTimestampOf</code>. See the <a href="#usingtimeuuid">section dedicated to these methods</a> for more detail.</li><li>“Float constants”#constants now support the exponent notation. In other words, <code>4.2E10</code> is now a valid floating point valu
e.</li></ul><h2 id="Versioning">Versioning</h2><p>Versioning of the CQL language adheres to the <a href="http://semver.org">Semantic Versioning</a> guidelines. Versions take the form X.Y.Z where X, Y, and Z are integer values representing major, minor, and patch level respectively. There is no correlation between Cassandra release versions and the CQL language version.</p><table><tr><th>version</th><th>description</th></tr><tr><td>Major </td><td>The major version <em>must</em> be bumped when backward incompatible changes are introduced. This should rarely occur.</td></tr><tr><td>Minor </td><td>Minor version increments occur when new, but backward compatible, functionality is introduced.</td></tr><tr><td>Patch </td><td>The patch version is incremented when bugs are fixed.</td></tr></table></body></html>
\ No newline at end of file