http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_ddl.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_ddl.html b/docs/build/html/topics/impala_ddl.html new file mode 100644 index 0000000..d7cf482 --- /dev/null +++ b/docs/build/html/topics/impala_ddl.html @@ -0,0 +1,141 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_langref_sql.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="ddl"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>DDL Statements</title></head><body id="ddl"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">DDL Statements</h1> + + + <div class="body conbody"> + + <p class="p"> + DDL refers to <span class="q">"Data Definition Language"</span>, a subset of SQL statements that change the structure of the + database schema in some way, typically by creating, deleting, or modifying schema objects such as databases, + tables, and views. Most Impala DDL statements start with the keywords <code class="ph codeph">CREATE</code>, + <code class="ph codeph">DROP</code>, or <code class="ph codeph">ALTER</code>. + </p> + + <p class="p"> + The Impala DDL statements are: + </p> + + <ul class="ul"> + <li class="li"> + <a class="xref" href="impala_alter_table.html#alter_table">ALTER TABLE Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_alter_view.html#alter_view">ALTER VIEW Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_compute_stats.html#compute_stats">COMPUTE STATS Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_create_database.html#create_database">CREATE DATABASE Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_create_function.html#create_function">CREATE FUNCTION Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_create_role.html#create_role">CREATE ROLE Statement (Impala 2.0 or higher only)</a> + </li> + + <li class="li"> + <a class="xref" href="impala_create_table.html#create_table">CREATE TABLE Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_create_view.html#create_view">CREATE VIEW Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_drop_database.html#drop_database">DROP DATABASE Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_drop_function.html#drop_function">DROP FUNCTION Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_drop_role.html#drop_role">DROP ROLE Statement (Impala 2.0 or higher only)</a> + </li> + + <li class="li"> + <a class="xref" href="impala_drop_table.html#drop_table">DROP TABLE Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_drop_view.html#drop_view">DROP VIEW Statement</a> + </li> + + <li class="li"> + <a class="xref" href="impala_grant.html#grant">GRANT Statement (Impala 2.0 or higher only)</a> + </li> + + <li class="li"> + <a class="xref" href="impala_revoke.html#revoke">REVOKE Statement (Impala 2.0 or higher only)</a> + </li> + </ul> + + <p class="p"> + After Impala executes a DDL command, information about available tables, columns, views, partitions, and so + on is automatically synchronized between all the Impala nodes in a cluster. (Prior to Impala 1.2, you had to + issue a <code class="ph codeph">REFRESH</code> or <code class="ph codeph">INVALIDATE METADATA</code> statement manually on the other + nodes to make them aware of the changes.) + </p> + + <p class="p"> + If the timing of metadata updates is significant, for example if you use round-robin scheduling where each + query could be issued through a different Impala node, you can enable the + <a class="xref" href="impala_sync_ddl.html#sync_ddl">SYNC_DDL</a> query option to make the DDL statement wait until + all nodes have been notified about the metadata changes. + </p> + + <p class="p"> + See <a class="xref" href="impala_s3.html#s3">Using Impala with the Amazon S3 Filesystem</a> for details about how Impala DDL statements interact with + tables and partitions stored in the Amazon S3 filesystem. + </p> + + <p class="p"> + Although the <code class="ph codeph">INSERT</code> statement is officially classified as a DML (data manipulation language) + statement, it also involves metadata changes that must be broadcast to all Impala nodes, and so is also + affected by the <code class="ph codeph">SYNC_DDL</code> query option. + </p> + + <p class="p"> + Because the <code class="ph codeph">SYNC_DDL</code> query option makes each DDL operation take longer than normal, you + might only enable it before the last DDL operation in a sequence. For example, if you are running a script + that issues multiple of DDL operations to set up an entire new schema, add several new partitions, and so on, + you might minimize the performance overhead by enabling the query option only before the last + <code class="ph codeph">CREATE</code>, <code class="ph codeph">DROP</code>, <code class="ph codeph">ALTER</code>, or <code class="ph codeph">INSERT</code> statement. + The script only finishes when all the relevant metadata changes are recognized by all the Impala nodes, so + you could connect to any node and issue queries through it. + </p> + + <p class="p"> + The classification of DDL, DML, and other statements is not necessarily the same between Impala and Hive. + Impala organizes these statements in a way intended to be familiar to people familiar with relational + databases or data warehouse products. Statements that modify the metastore database, such as <code class="ph codeph">COMPUTE + STATS</code>, are classified as DDL. Statements that only query the metastore database, such as + <code class="ph codeph">SHOW</code> or <code class="ph codeph">DESCRIBE</code>, are put into a separate category of utility statements. + </p> + + <div class="note note note_note"><span class="note__title notetitle">Note:</span> + The query types shown in the Impala debug web user interface might not match exactly the categories listed + here. For example, currently the <code class="ph codeph">USE</code> statement is shown as DDL in the debug web UI. The + query types shown in the debug web UI are subject to change, for improved consistency. + </div> + + <p class="p"> + <strong class="ph b">Related information:</strong> + </p> + + <p class="p"> + The other major classifications of SQL statements are data manipulation language (see + <a class="xref" href="impala_dml.html#dml">DML Statements</a>) and queries (see <a class="xref" href="impala_select.html#select">SELECT Statement</a>). + </p> + </div> +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_langref_sql.html">Impala SQL Statements</a></div></div></nav></article></main></body></html> \ No newline at end of file
http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_debug_action.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_debug_action.html b/docs/build/html/topics/impala_debug_action.html new file mode 100644 index 0000000..ce4ef7c --- /dev/null +++ b/docs/build/html/topics/impala_debug_action.html @@ -0,0 +1,24 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_query_options.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="debug_action"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>DEBUG_ACTION Query Option</title></head><body id="debug_action"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">DEBUG_ACTION Query Option</h1> + + + + <div class="body conbody"> + + <p class="p"> + + Introduces artificial problem conditions within queries. For internal debugging and troubleshooting. + </p> + + <p class="p"> + <strong class="ph b">Type:</strong> <code class="ph codeph">STRING</code> + </p> + + <p class="p"> + <strong class="ph b">Default:</strong> empty string + </p> + </div> +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_query_options.html">Query Options for the SET Statement</a></div></div></nav></article></main></body></html> \ No newline at end of file http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_decimal.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_decimal.html b/docs/build/html/topics/impala_decimal.html new file mode 100644 index 0000000..8cec53e --- /dev/null +++ b/docs/build/html/topics/impala_decimal.html @@ -0,0 +1,826 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_datatypes.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="decimal"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>DECIMAL Data Type (Impala 1.4 or higher only)</title></head><body id="decimal"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">DECIMAL Data Type (<span class="keyword">Impala 1.4</span> or higher only)</h1> + + + + <div class="body conbody"> + + <p class="p"> + A numeric data type with fixed scale and precision, used in <code class="ph codeph">CREATE TABLE</code> and <code class="ph codeph">ALTER + TABLE</code> statements. Suitable for financial and other arithmetic calculations where the imprecise + representation and rounding behavior of <code class="ph codeph">FLOAT</code> and <code class="ph codeph">DOUBLE</code> make those types + impractical. + </p> + + <p class="p"> + <strong class="ph b">Syntax:</strong> + </p> + + <p class="p"> + In the column definition of a <code class="ph codeph">CREATE TABLE</code> statement: + </p> + +<pre class="pre codeblock"><code><var class="keyword varname">column_name</var> DECIMAL[(<var class="keyword varname">precision</var>[,<var class="keyword varname">scale</var>])]</code></pre> + + <p class="p"> + <code class="ph codeph">DECIMAL</code> with no precision or scale values is equivalent to <code class="ph codeph">DECIMAL(9,0)</code>. + </p> + + <p class="p"> + <strong class="ph b">Precision and Scale:</strong> + </p> + + <p class="p"> + <var class="keyword varname">precision</var> represents the total number of digits that can be represented by the column, + regardless of the location of the decimal point. This value must be between 1 and 38. For example, + representing integer values up to 9999, and floating-point values up to 99.99, both require a precision of 4. + You can also represent corresponding negative values, without any change in the precision. For example, the + range -9999 to 9999 still only requires a precision of 4. + </p> + + <p class="p"> + <var class="keyword varname">scale</var> represents the number of fractional digits. This value must be less than or equal to + <var class="keyword varname">precision</var>. A scale of 0 produces integral values, with no fractional part. If precision + and scale are equal, all the digits come after the decimal point, making all the values between 0 and + 0.999... or 0 and -0.999... + </p> + + <p class="p"> + When <var class="keyword varname">precision</var> and <var class="keyword varname">scale</var> are omitted, a <code class="ph codeph">DECIMAL</code> value + is treated as <code class="ph codeph">DECIMAL(9,0)</code>, that is, an integer value ranging from + <code class="ph codeph">-999,999,999</code> to <code class="ph codeph">999,999,999</code>. This is the largest <code class="ph codeph">DECIMAL</code> + value that can still be represented in 4 bytes. If precision is specified but scale is omitted, Impala uses a + value of zero for the scale. + </p> + + <p class="p"> + Both <var class="keyword varname">precision</var> and <var class="keyword varname">scale</var> must be specified as integer literals, not any + other kind of constant expressions. + </p> + + <p class="p"> + To check the precision or scale for arbitrary values, you can call the + <a class="xref" href="impala_math_functions.html#math_functions"><code class="ph codeph">precision()</code> and + <code class="ph codeph">scale()</code> built-in functions</a>. For example, you might use these values to figure out how + many characters are required for various fields in a report, or to understand the rounding characteristics of + a formula as applied to a particular <code class="ph codeph">DECIMAL</code> column. + </p> + + <p class="p"> + <strong class="ph b">Range:</strong> + </p> + + <p class="p"> + The maximum precision value is 38. Thus, the largest integral value is represented by + <code class="ph codeph">DECIMAL(38,0)</code> (999... with 9 repeated 38 times). The most precise fractional value (between + 0 and 1, or 0 and -1) is represented by <code class="ph codeph">DECIMAL(38,38)</code>, with 38 digits to the right of the + decimal point. The value closest to 0 would be .0000...1 (37 zeros and the final 1). The value closest to 1 + would be .999... (9 repeated 38 times). + </p> + + <p class="p"> + For a given precision and scale, the range of <code class="ph codeph">DECIMAL</code> values is the same in the positive and + negative directions. For example, <code class="ph codeph">DECIMAL(4,2)</code> can represent from -99.99 to 99.99. This is + different from other integral numeric types where the positive and negative bounds differ slightly. + </p> + + <p class="p"> + When you use <code class="ph codeph">DECIMAL</code> values in arithmetic expressions, the precision and scale of the result + value are determined as follows: + </p> + + <ul class="ul"> + <li class="li"> + <p class="p"> + For addition and subtraction, the precision and scale are based on the maximum possible result, that is, + if all the digits of the input values were 9s and the absolute values were added together. + </p> + + + </li> + + <li class="li"> + <p class="p"> + For multiplication, the precision is the sum of the precisions of the input values. The scale is the sum + of the scales of the input values. + </p> + </li> + + + + <li class="li"> + <p class="p"> + For division, Impala sets the precision and scale to values large enough to represent the whole and + fractional parts of the result. + </p> + </li> + + <li class="li"> + <p class="p"> + For <code class="ph codeph">UNION</code>, the scale is the larger of the scales of the input values, and the precision + is increased if necessary to accommodate any additional fractional digits. If the same input value has + the largest precision and the largest scale, the result value has the same precision and scale. If one + value has a larger precision but smaller scale, the scale of the result value is increased. For example, + <code class="ph codeph">DECIMAL(20,2) UNION DECIMAL(8,6)</code> produces a result of type + <code class="ph codeph">DECIMAL(24,6)</code>. The extra 4 fractional digits of scale (6-2) are accommodated by + extending the precision by the same amount (20+4). + </p> + </li> + + <li class="li"> + <p class="p"> + To doublecheck, you can always call the <code class="ph codeph">PRECISION()</code> and <code class="ph codeph">SCALE()</code> + functions on the results of an arithmetic expression to see the relevant values, or use a <code class="ph codeph">CREATE + TABLE AS SELECT</code> statement to define a column based on the return type of the expression. + </p> + </li> + </ul> + + <p class="p"> + <strong class="ph b">Compatibility:</strong> + </p> + + <ul class="ul"> + <li class="li"> + Using the <code class="ph codeph">DECIMAL</code> type is only supported under <span class="keyword">Impala 1.4</span> and higher. + </li> + + <li class="li"> + Use the <code class="ph codeph">DECIMAL</code> data type in Impala for applications where you used the + <code class="ph codeph">NUMBER</code> data type in Oracle. The Impala <code class="ph codeph">DECIMAL</code> type does not support the + Oracle idioms of <code class="ph codeph">*</code> for scale or negative values for precision. + </li> + </ul> + + <p class="p"> + <strong class="ph b">Conversions and casting:</strong> + </p> + + <p class="p"> + <span class="ph">Casting an integer or floating-point value <code class="ph codeph">N</code> to + <code class="ph codeph">TIMESTAMP</code> produces a value that is <code class="ph codeph">N</code> seconds past the start of the epoch + date (January 1, 1970). By default, the result value represents a date and time in the UTC time zone. + If the setting <code class="ph codeph">-use_local_tz_for_unix_timestamp_conversions=true</code> is in effect, + the resulting <code class="ph codeph">TIMESTAMP</code> represents a date and time in the local time zone.</span> + </p> + + <p class="p"> + Impala automatically converts between <code class="ph codeph">DECIMAL</code> and other numeric types where possible. A + <code class="ph codeph">DECIMAL</code> with zero scale is converted to or from the smallest appropriate integral type. A + <code class="ph codeph">DECIMAL</code> with a fractional part is automatically converted to or from the smallest + appropriate floating-point type. If the destination type does not have sufficient precision or scale to hold + all possible values of the source type, Impala raises an error and does not convert the value. + </p> + + <p class="p"> + For example, these statements show how expressions of <code class="ph codeph">DECIMAL</code> and other types are reconciled + to the same type in the context of <code class="ph codeph">UNION</code> queries and <code class="ph codeph">INSERT</code> statements: + </p> + +<pre class="pre codeblock"><code>[localhost:21000] > select cast(1 as int) as x union select cast(1.5 as decimal(9,4)) as x; ++----------------+ +| x | ++----------------+ +| 1.5000 | +| 1.0000 | ++----------------+ +[localhost:21000] > create table int_vs_decimal as select cast(1 as int) as x union select cast(1.5 as decimal(9,4)) as x; ++-------------------+ +| summary | ++-------------------+ +| Inserted 2 row(s) | ++-------------------+ +[localhost:21000] > desc int_vs_decimal; ++------+---------------+---------+ +| name | type | comment | ++------+---------------+---------+ +| x | decimal(14,4) | | ++------+---------------+---------+ + +</code></pre> + + <p class="p"> + To avoid potential conversion errors, you can use <code class="ph codeph">CAST()</code> to convert <code class="ph codeph">DECIMAL</code> + values to <code class="ph codeph">FLOAT</code>, <code class="ph codeph">TINYINT</code>, <code class="ph codeph">SMALLINT</code>, <code class="ph codeph">INT</code>, + <code class="ph codeph">BIGINT</code>, <code class="ph codeph">STRING</code>, <code class="ph codeph">TIMESTAMP</code>, or <code class="ph codeph">BOOLEAN</code>. + You can use exponential notation in <code class="ph codeph">DECIMAL</code> literals or when casting from + <code class="ph codeph">STRING</code>, for example <code class="ph codeph">1.0e6</code> to represent one million. + </p> + + <p class="p"> + If you cast a value with more fractional digits than the scale of the destination type, any extra fractional + digits are truncated (not rounded). Casting a value to a target type with not enough precision produces a + result of <code class="ph codeph">NULL</code> and displays a runtime warning. + </p> + +<pre class="pre codeblock"><code>[localhost:21000] > select cast(1.239 as decimal(3,2)); ++-----------------------------+ +| cast(1.239 as decimal(3,2)) | ++-----------------------------+ +| 1.23 | ++-----------------------------+ +[localhost:21000] > select cast(1234 as decimal(3)); ++----------------------------+ +| cast(1234 as decimal(3,0)) | ++----------------------------+ +| NULL | ++----------------------------+ +WARNINGS: Expression overflowed, returning NULL + +</code></pre> + + <p class="p"> + When you specify integer literals, for example in <code class="ph codeph">INSERT ... VALUES</code> statements or arithmetic + expressions, those numbers are interpreted as the smallest applicable integer type. You must use + <code class="ph codeph">CAST()</code> calls for some combinations of integer literals and <code class="ph codeph">DECIMAL</code> + precision. For example, <code class="ph codeph">INT</code> has a maximum value that is 10 digits long, + <code class="ph codeph">TINYINT</code> has a maximum value that is 3 digits long, and so on. If you specify a value such as + 123456 to go into a <code class="ph codeph">DECIMAL</code> column, Impala checks if the column has enough precision to + represent the largest value of that integer type, and raises an error if not. Therefore, use an expression + like <code class="ph codeph">CAST(123456 TO DECIMAL(9,0))</code> for <code class="ph codeph">DECIMAL</code> columns with precision 9 or + less, <code class="ph codeph">CAST(50 TO DECIMAL(2,0))</code> for <code class="ph codeph">DECIMAL</code> columns with precision 2 or + less, and so on. For <code class="ph codeph">DECIMAL</code> columns with precision 10 or greater, Impala automatically + interprets the value as the correct <code class="ph codeph">DECIMAL</code> type; however, because + <code class="ph codeph">DECIMAL(10)</code> requires 8 bytes of storage while <code class="ph codeph">DECIMAL(9)</code> requires only 4 + bytes, only use precision of 10 or higher when actually needed. + </p> + +<pre class="pre codeblock"><code>[localhost:21000] > create table decimals_9_0 (x decimal); +[localhost:21000] > insert into decimals_9_0 values (1), (2), (4), (8), (16), (1024), (32768), (65536), (1000000); +ERROR: AnalysisException: Possible loss of precision for target table 'decimal_testing.decimals_9_0'. +Expression '1' (type: INT) would need to be cast to DECIMAL(9,0) for column 'x' +[localhost:21000] > insert into decimals_9_0 values (cast(1 as decimal)), (cast(2 as decimal)), (cast(4 as decimal)), (cast(8 as decimal)), (cast(16 as decimal)), (cast(1024 as decimal)), (cast(32768 as decimal)), (cast(65536 as decimal)), (cast(1000000 as decimal)); + +[localhost:21000] > create table decimals_10_0 (x decimal(10,0)); +[localhost:21000] > insert into decimals_10_0 values (1), (2), (4), (8), (16), (1024), (32768), (65536), (1000000); + +</code></pre> + + <p class="p"> + Be aware that in memory and for binary file formats such as Parquet or Avro, <code class="ph codeph">DECIMAL(10)</code> or + higher consumes 8 bytes while <code class="ph codeph">DECIMAL(9)</code> (the default for <code class="ph codeph">DECIMAL</code>) or lower + consumes 4 bytes. Therefore, to conserve space in large tables, use the smallest-precision + <code class="ph codeph">DECIMAL</code> type that is appropriate and <code class="ph codeph">CAST()</code> literal values where necessary, + rather than declaring <code class="ph codeph">DECIMAL</code> columns with high precision for convenience. + </p> + + <p class="p"> + To represent a very large or precise <code class="ph codeph">DECIMAL</code> value as a literal, for example one that + contains more digits than can be represented by a <code class="ph codeph">BIGINT</code> literal, use a quoted string or a + floating-point value for the number, and <code class="ph codeph">CAST()</code> to the desired <code class="ph codeph">DECIMAL</code> + type: + </p> + +<pre class="pre codeblock"><code>insert into decimals_38_5 values (1), (2), (4), (8), (16), (1024), (32768), (65536), (1000000), + (cast("999999999999999999999999999999" as decimal(38,5))), + (cast(999999999999999999999999999999. as decimal(38,5))); +</code></pre> + + <ul class="ul"> + <li class="li"> + <p class="p"> The result of the <code class="ph codeph">SUM()</code> aggregate function on + <code class="ph codeph">DECIMAL</code> values is promoted to a precision of 38, + with the same precision as the underlying column. Thus, the result can + represent the largest possible value at that particular precision. </p> + </li> + + <li class="li"> + <p class="p"> + <code class="ph codeph">STRING</code> columns, literals, or expressions can be converted to <code class="ph codeph">DECIMAL</code> as + long as the overall number of digits and digits to the right of the decimal point fit within the + specified precision and scale for the declared <code class="ph codeph">DECIMAL</code> type. By default, a + <code class="ph codeph">DECIMAL</code> value with no specified scale or precision can hold a maximum of 9 digits of an + integer value. If there are more digits in the string value than are allowed by the + <code class="ph codeph">DECIMAL</code> scale and precision, the result is <code class="ph codeph">NULL</code>. + </p> + <p class="p"> + The following examples demonstrate how <code class="ph codeph">STRING</code> values with integer and fractional parts + are represented when converted to <code class="ph codeph">DECIMAL</code>. If the scale is 0, the number is treated + as an integer value with a maximum of <var class="keyword varname">precision</var> digits. If the precision is greater than + 0, the scale must be increased to account for the digits both to the left and right of the decimal point. + As the precision increases, output values are printed with additional trailing zeros after the decimal + point if needed. Any trailing zeros after the decimal point in the <code class="ph codeph">STRING</code> value must fit + within the number of digits specified by the precision. + </p> +<pre class="pre codeblock"><code>[localhost:21000] > select cast('100' as decimal); -- Small integer value fits within 9 digits of scale. ++-----------------------------+ +| cast('100' as decimal(9,0)) | ++-----------------------------+ +| 100 | ++-----------------------------+ +[localhost:21000] > select cast('100' as decimal(3,0)); -- Small integer value fits within 3 digits of scale. ++-----------------------------+ +| cast('100' as decimal(3,0)) | ++-----------------------------+ +| 100 | ++-----------------------------+ +[localhost:21000] > select cast('100' as decimal(2,0)); -- 2 digits of scale is not enough! ++-----------------------------+ +| cast('100' as decimal(2,0)) | ++-----------------------------+ +| NULL | ++-----------------------------+ +[localhost:21000] > select cast('100' as decimal(3,1)); -- (3,1) = 2 digits left of the decimal point, 1 to the right. Not enough. ++-----------------------------+ +| cast('100' as decimal(3,1)) | ++-----------------------------+ +| NULL | ++-----------------------------+ +[localhost:21000] > select cast('100' as decimal(4,1)); -- 4 digits total, 1 to the right of the decimal point. ++-----------------------------+ +| cast('100' as decimal(4,1)) | ++-----------------------------+ +| 100.0 | ++-----------------------------+ +[localhost:21000] > select cast('98.6' as decimal(3,1)); -- (3,1) can hold a 3 digit number with 1 fractional digit. ++------------------------------+ +| cast('98.6' as decimal(3,1)) | ++------------------------------+ +| 98.6 | ++------------------------------+ +[localhost:21000] > select cast('98.6' as decimal(15,1)); -- Larger scale allows bigger numbers but still only 1 fractional digit. ++-------------------------------+ +| cast('98.6' as decimal(15,1)) | ++-------------------------------+ +| 98.6 | ++-------------------------------+ +[localhost:21000] > select cast('98.6' as decimal(15,5)); -- Larger precision allows more fractional digits, outputs trailing zeros. ++-------------------------------+ +| cast('98.6' as decimal(15,5)) | ++-------------------------------+ +| 98.60000 | ++-------------------------------+ +[localhost:21000] > select cast('98.60000' as decimal(15,1)); -- Trailing zeros in the string must fit within 'scale' digits (1 in this case). ++-----------------------------------+ +| cast('98.60000' as decimal(15,1)) | ++-----------------------------------+ +| NULL | ++-----------------------------------+ + +</code></pre> + </li> + + <li class="li"> + Most built-in arithmetic functions such as <code class="ph codeph">SIN()</code> and <code class="ph codeph">COS()</code> continue to + accept only <code class="ph codeph">DOUBLE</code> values because they are so commonly used in scientific context for + calculations of IEEE 954-compliant values. The built-in functions that accept and return + <code class="ph codeph">DECIMAL</code> are: + + + <ul class="ul"> + <li class="li"> + <code class="ph codeph">ABS()</code> + </li> + + <li class="li"> + <code class="ph codeph">CEIL()</code> + </li> + + <li class="li"> + <code class="ph codeph">COALESCE()</code> + </li> + + <li class="li"> + <code class="ph codeph">FLOOR()</code> + </li> + + <li class="li"> + <code class="ph codeph">FNV_HASH()</code> + </li> + + <li class="li"> + <code class="ph codeph">GREATEST()</code> + </li> + + <li class="li"> + <code class="ph codeph">IF()</code> + </li> + + <li class="li"> + <code class="ph codeph">ISNULL()</code> + </li> + + <li class="li"> + <code class="ph codeph">LEAST()</code> + </li> + + <li class="li"> + <code class="ph codeph">NEGATIVE()</code> + </li> + + <li class="li"> + <code class="ph codeph">NULLIF()</code> + </li> + + <li class="li"> + <code class="ph codeph">POSITIVE()</code> + </li> + + <li class="li"> + <code class="ph codeph">PRECISION()</code> + </li> + + <li class="li"> + <code class="ph codeph">ROUND()</code> + </li> + + <li class="li"> + <code class="ph codeph">SCALE()</code> + </li> + + <li class="li"> + <code class="ph codeph">TRUNCATE()</code> + </li> + + <li class="li"> + <code class="ph codeph">ZEROIFNULL()</code> + </li> + </ul> + See <a class="xref" href="impala_functions.html#builtins">Impala Built-In Functions</a> for details. + </li> + + <li class="li"> + <p class="p"> + <code class="ph codeph">BIGINT</code>, <code class="ph codeph">INT</code>, <code class="ph codeph">SMALLINT</code>, and <code class="ph codeph">TINYINT</code> + values can all be cast to <code class="ph codeph">DECIMAL</code>. The number of digits to the left of the decimal point + in the <code class="ph codeph">DECIMAL</code> type must be sufficient to hold the largest value of the corresponding + integer type. Note that integer literals are treated as the smallest appropriate integer type, meaning + there is sometimes a range of values that require one more digit of <code class="ph codeph">DECIMAL</code> scale than + you might expect. For integer values, the precision of the <code class="ph codeph">DECIMAL</code> type can be zero; if + the precision is greater than zero, remember to increase the scale value by an equivalent amount to hold + the required number of digits to the left of the decimal point. + </p> + <p class="p"> + The following examples show how different integer types are converted to <code class="ph codeph">DECIMAL</code>. + </p> + +<pre class="pre codeblock"><code>[localhost:21000] > select cast(1 as decimal(1,0)); ++-------------------------+ +| cast(1 as decimal(1,0)) | ++-------------------------+ +| 1 | ++-------------------------+ +[localhost:21000] > select cast(9 as decimal(1,0)); ++-------------------------+ +| cast(9 as decimal(1,0)) | ++-------------------------+ +| 9 | ++-------------------------+ +[localhost:21000] > select cast(10 as decimal(1,0)); ++--------------------------+ +| cast(10 as decimal(1,0)) | ++--------------------------+ +| 10 | ++--------------------------+ +[localhost:21000] > select cast(10 as decimal(1,1)); ++--------------------------+ +| cast(10 as decimal(1,1)) | ++--------------------------+ +| 10.0 | ++--------------------------+ +[localhost:21000] > select cast(100 as decimal(1,1)); ++---------------------------+ +| cast(100 as decimal(1,1)) | ++---------------------------+ +| 100.0 | ++---------------------------+ +[localhost:21000] > select cast(1000 as decimal(1,1)); ++----------------------------+ +| cast(1000 as decimal(1,1)) | ++----------------------------+ +| 1000.0 | ++----------------------------+ + +</code></pre> + </li> + + <li class="li"> + <p class="p"> + When a <code class="ph codeph">DECIMAL</code> value is converted to any of the integer types, any fractional part is + truncated (that is, rounded towards zero): + </p> +<pre class="pre codeblock"><code>[localhost:21000] > create table num_dec_days (x decimal(4,1)); +[localhost:21000] > insert into num_dec_days values (1), (2), (cast(4.5 as decimal(4,1))); +[localhost:21000] > insert into num_dec_days values (cast(0.1 as decimal(4,1))), (cast(.9 as decimal(4,1))), (cast(9.1 as decimal(4,1))), (cast(9.9 as decimal(4,1))); +[localhost:21000] > select cast(x as int) from num_dec_days; ++----------------+ +| cast(x as int) | ++----------------+ +| 1 | +| 2 | +| 4 | +| 0 | +| 0 | +| 9 | +| 9 | ++----------------+ + +</code></pre> + </li> + + <li class="li"> + <p class="p"> + You cannot directly cast <code class="ph codeph">TIMESTAMP</code> or <code class="ph codeph">BOOLEAN</code> values to or from + <code class="ph codeph">DECIMAL</code> values. You can turn a <code class="ph codeph">DECIMAL</code> value into a time-related + representation using a two-step process, by converting it to an integer value and then using that result + in a call to a date and time function such as <code class="ph codeph">from_unixtime()</code>. + </p> +<pre class="pre codeblock"><code>[localhost:21000] > select from_unixtime(cast(cast(1000.0 as decimal) as bigint)); ++-------------------------------------------------------------+ +| from_unixtime(cast(cast(1000.0 as decimal(9,0)) as bigint)) | ++-------------------------------------------------------------+ +| 1970-01-01 00:16:40 | ++-------------------------------------------------------------+ +[localhost:21000] > select now() + interval cast(x as int) days from num_dec_days; -- x is a DECIMAL column. + +[localhost:21000] > create table num_dec_days (x decimal(4,1)); +[localhost:21000] > insert into num_dec_days values (1), (2), (cast(4.5 as decimal(4,1))); +[localhost:21000] > select now() + interval cast(x as int) days from num_dec_days; -- The 4.5 value is truncated to 4 and becomes '4 days'. ++--------------------------------------+ +| now() + interval cast(x as int) days | ++--------------------------------------+ +| 2014-05-13 23:11:55.163284000 | +| 2014-05-14 23:11:55.163284000 | +| 2014-05-16 23:11:55.163284000 | ++--------------------------------------+ + +</code></pre> + </li> + + <li class="li"> + <p class="p"> + Because values in <code class="ph codeph">INSERT</code> statements are checked rigorously for type compatibility, be + prepared to use <code class="ph codeph">CAST()</code> function calls around literals, column references, or other + expressions that you are inserting into a <code class="ph codeph">DECIMAL</code> column. + </p> + </li> + </ul> + + <p class="p"> + <strong class="ph b">NULL considerations:</strong> Casting any non-numeric value to this type produces a <code class="ph codeph">NULL</code> + value. + </p> + + <p class="p"> + <strong class="ph b">DECIMAL differences from integer and floating-point types:</strong> + </p> + + <p class="p"> + With the <code class="ph codeph">DECIMAL</code> type, you are concerned with the number of overall digits of a number + rather than powers of 2 (as in <code class="ph codeph">TINYINT</code>, <code class="ph codeph">SMALLINT</code>, and so on). Therefore, + the limits with integral values of <code class="ph codeph">DECIMAL</code> types fall around 99, 999, 9999, and so on rather + than 32767, 65535, 2 + <sup class="ph sup">32</sup> + -1, and so on. For fractional values, you do not need to account for imprecise representation of the + fractional part according to the IEEE-954 standard (as in <code class="ph codeph">FLOAT</code> and + <code class="ph codeph">DOUBLE</code>). Therefore, when you insert a fractional value into a <code class="ph codeph">DECIMAL</code> + column, you can compare, sum, query, <code class="ph codeph">GROUP BY</code>, and so on that column and get back the + original values rather than some <span class="q">"close but not identical"</span> value. + </p> + + <p class="p"> + <code class="ph codeph">FLOAT</code> and <code class="ph codeph">DOUBLE</code> can cause problems or unexpected behavior due to inability + to precisely represent certain fractional values, for example dollar and cents values for currency. You might + find output values slightly different than you inserted, equality tests that do not match precisely, or + unexpected values for <code class="ph codeph">GROUP BY</code> columns. <code class="ph codeph">DECIMAL</code> can help reduce unexpected + behavior and rounding errors, at the expense of some performance overhead for assignments and comparisons. + </p> + + <div class="p"> + <strong class="ph b">Literals and expressions:</strong> + <ul class="ul"> + <li class="li"> + <p class="p"> + When you use an integer literal such as <code class="ph codeph">1</code> or <code class="ph codeph">999</code> in a SQL statement, + depending on the context, Impala will treat it as either the smallest appropriate + <code class="ph codeph">DECIMAL</code> type, or the smallest integer type (<code class="ph codeph">TINYINT</code>, + <code class="ph codeph">SMALLINT</code>, <code class="ph codeph">INT</code>, or <code class="ph codeph">BIGINT</code>). To minimize memory usage, + Impala prefers to treat the literal as the smallest appropriate integer type. + </p> + </li> + + <li class="li"> + <p class="p"> + When you use a floating-point literal such as <code class="ph codeph">1.1</code> or <code class="ph codeph">999.44</code> in a SQL + statement, depending on the context, Impala will treat it as either the smallest appropriate + <code class="ph codeph">DECIMAL</code> type, or the smallest floating-point type (<code class="ph codeph">FLOAT</code> or + <code class="ph codeph">DOUBLE</code>). To avoid loss of accuracy, Impala prefers to treat the literal as a + <code class="ph codeph">DECIMAL</code>. + </p> + </li> + </ul> + </div> + + <p class="p"> + <strong class="ph b">Storage considerations:</strong> + </p> + + <ul class="ul"> + <li class="li"> + Only the precision determines the storage size for <code class="ph codeph">DECIMAL</code> values; the scale setting has + no effect on the storage size. + </li> + + <li class="li"> + Text, RCFile, and SequenceFile tables all use ASCII-based formats. In these text-based file formats, + leading zeros are not stored, but trailing zeros are stored. In these tables, each <code class="ph codeph">DECIMAL</code> + value takes up as many bytes as there are digits in the value, plus an extra byte if the decimal point is + present and an extra byte for negative values. Once the values are loaded into memory, they are represented + in 4, 8, or 16 bytes as described in the following list items. The on-disk representation varies depending + on the file format of the table. + </li> + + + + <li class="li"> + Parquet and Avro tables use binary formats, In these tables, Impala stores each value in as few bytes as + possible + + depending on the precision specified for the <code class="ph codeph">DECIMAL</code> column. + <ul class="ul"> + <li class="li"> + In memory, <code class="ph codeph">DECIMAL</code> values with precision of 9 or less are stored in 4 bytes. + </li> + + <li class="li"> + In memory, <code class="ph codeph">DECIMAL</code> values with precision of 10 through 18 are stored in 8 bytes. + </li> + + <li class="li"> + In memory, <code class="ph codeph">DECIMAL</code> values with precision greater than 18 are stored in 16 bytes. + </li> + </ul> + </li> + </ul> + + <p class="p"> + <strong class="ph b">File format considerations:</strong> + </p> + + <ul class="ul"> + <li class="li"> + The <code class="ph codeph">DECIMAL</code> data type can be stored in any of the file formats supported by Impala, as + described in <a class="xref" href="impala_file_formats.html#file_formats">How Impala Works with Hadoop File Formats</a>. Impala only writes to tables that use the + Parquet and text formats, so those formats are the focus for file format compatibility. + </li> + + <li class="li"> + Impala can query Avro, RCFile, or SequenceFile tables containing <code class="ph codeph">DECIMAL</code> columns, created + by other Hadoop components. + </li> + + <li class="li"> + You can use <code class="ph codeph">DECIMAL</code> columns in Impala tables that are mapped to HBase tables. Impala can + query and insert into such tables. + </li> + + <li class="li"> + Text, RCFile, and SequenceFile tables all use ASCII-based formats. In these tables, each + <code class="ph codeph">DECIMAL</code> value takes up as many bytes as there are digits in the value, plus an extra byte + if the decimal point is present. The binary format of Parquet or Avro files offers more compact storage for + <code class="ph codeph">DECIMAL</code> columns. + </li> + + <li class="li"> + Parquet and Avro tables use binary formats, In these tables, Impala stores each value in 4, 8, or 16 bytes + depending on the precision specified for the <code class="ph codeph">DECIMAL</code> column. + </li> + + </ul> + + <p class="p"> + <strong class="ph b">UDF considerations:</strong> When writing a C++ UDF, use the <code class="ph codeph">DecimalVal</code> data type defined in + <span class="ph filepath">/usr/include/impala_udf/udf.h</span>. + </p> + + <p class="p"> + <strong class="ph b">Partitioning:</strong> + </p> + + <p class="p"> + You can use a <code class="ph codeph">DECIMAL</code> column as a partition key. Doing so provides a better match between + the partition key values and the HDFS directory names than using a <code class="ph codeph">DOUBLE</code> or + <code class="ph codeph">FLOAT</code> partitioning column: + </p> + + <p class="p"> + <strong class="ph b">Schema evolution considerations:</strong> + </p> + + <ul class="ul"> + <li class="li"> + For text-based formats (text, RCFile, and SequenceFile tables), you can issue an <code class="ph codeph">ALTER TABLE ... + REPLACE COLUMNS</code> statement to change the precision and scale of an existing + <code class="ph codeph">DECIMAL</code> column. As long as the values in the column fit within the new precision and + scale, they are returned correctly by a query. Any values that do not fit within the new precision and + scale are returned as <code class="ph codeph">NULL</code>, and Impala reports the conversion error. Leading zeros do not + count against the precision value, but trailing zeros after the decimal point do. +<pre class="pre codeblock"><code>[localhost:21000] > create table text_decimals (x string); +[localhost:21000] > insert into text_decimals values ("1"), ("2"), ("99.99"), ("1.234"), ("000001"), ("1.000000000"); +[localhost:21000] > select * from text_decimals; ++-------------+ +| x | ++-------------+ +| 1 | +| 2 | +| 99.99 | +| 1.234 | +| 000001 | +| 1.000000000 | ++-------------+ +[localhost:21000] > alter table text_decimals replace columns (x decimal(4,2)); +[localhost:21000] > select * from text_decimals; ++-------+ +| x | ++-------+ +| 1.00 | +| 2.00 | +| 99.99 | +| NULL | +| 1.00 | +| NULL | ++-------+ +ERRORS: +Backend 0:Error converting column: 0 TO DECIMAL(4, 2) (Data is: 1.234) +file: hdfs://127.0.0.1:8020/user/hive/warehouse/decimal_testing.db/text_decimals/634d4bd3aa0 +e8420-b4b13bab7f1be787_56794587_data.0 +record: 1.234 +Error converting column: 0 TO DECIMAL(4, 2) (Data is: 1.000000000) +file: hdfs://127.0.0.1:8020/user/hive/warehouse/decimal_testing.db/text_decimals/cd40dc68e20 +c565a-cc4bd86c724c96ba_311873428_data.0 +record: 1.000000000 + +</code></pre> + </li> + + <li class="li"> + For binary formats (Parquet and Avro tables), although an <code class="ph codeph">ALTER TABLE ... REPLACE COLUMNS</code> + statement that changes the precision or scale of a <code class="ph codeph">DECIMAL</code> column succeeds, any subsequent + attempt to query the changed column results in a fatal error. (The other columns can still be queried + successfully.) This is because the metadata about the columns is stored in the data files themselves, and + <code class="ph codeph">ALTER TABLE</code> does not actually make any updates to the data files. If the metadata in the + data files disagrees with the metadata in the metastore database, Impala cancels the query. + </li> + </ul> + + <p class="p"> + <strong class="ph b">Examples:</strong> + </p> + +<pre class="pre codeblock"><code>CREATE TABLE t1 (x DECIMAL, y DECIMAL(5,2), z DECIMAL(25,0)); +INSERT INTO t1 VALUES (5, 99.44, 123456), (300, 6.7, 999999999); +SELECT x+y, ROUND(y,1), z/98.6 FROM t1; +SELECT CAST(1000.5 AS DECIMAL); +</code></pre> + + + + <p class="p"> + <strong class="ph b">HBase considerations:</strong> This data type is fully compatible with HBase tables. + </p> + + <p class="p"> + <strong class="ph b">Parquet considerations:</strong> This type is fully compatible with Parquet tables. + </p> + + <p class="p"> + <strong class="ph b">Text table considerations:</strong> Values of this type are potentially larger in text tables than in tables + using Parquet or other binary formats. + </p> + + + + + + + + <p class="p"> + <strong class="ph b">Column statistics considerations:</strong> Because this type has a fixed size, the maximum and average size + fields are always filled in for column statistics, even before you run the <code class="ph codeph">COMPUTE STATS</code> + statement. + </p> + + <p class="p"> + <strong class="ph b">Kudu considerations:</strong> + </p> + <p class="p"> + Currently, the data types <code class="ph codeph">DECIMAL</code>, <code class="ph codeph">TIMESTAMP</code>, <code class="ph codeph">CHAR</code>, <code class="ph codeph">VARCHAR</code>, + <code class="ph codeph">ARRAY</code>, <code class="ph codeph">MAP</code>, and <code class="ph codeph">STRUCT</code> cannot be used with Kudu tables. + </p> + + <p class="p"> + <strong class="ph b">Related information:</strong> + </p> + + <p class="p"> + <a class="xref" href="impala_literals.html#numeric_literals">Numeric Literals</a>, <a class="xref" href="impala_tinyint.html#tinyint">TINYINT Data Type</a>, + <a class="xref" href="impala_smallint.html#smallint">SMALLINT Data Type</a>, <a class="xref" href="impala_int.html#int">INT Data Type</a>, + <a class="xref" href="impala_bigint.html#bigint">BIGINT Data Type</a>, <a class="xref" href="impala_decimal.html#decimal">DECIMAL Data Type (Impala 1.4 or higher only)</a>, + <a class="xref" href="impala_math_functions.html#math_functions">Impala Mathematical Functions</a> (especially <code class="ph codeph">PRECISION()</code> and + <code class="ph codeph">SCALE()</code>) + </p> + </div> +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_datatypes.html">Data Types</a></div></div></nav></article></main></body></html> \ No newline at end of file http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_default_order_by_limit.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_default_order_by_limit.html b/docs/build/html/topics/impala_default_order_by_limit.html new file mode 100644 index 0000000..334253f --- /dev/null +++ b/docs/build/html/topics/impala_default_order_by_limit.html @@ -0,0 +1,33 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_query_options.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="default_order_by_limit"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>DEFAULT_ORDER_BY_LIMIT Query Option</title></head><body id="default_order_by_limit"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">DEFAULT_ORDER_BY_LIMIT Query Option</h1> + + + + <div class="body conbody"> + + <p class="p"> + Now that the <code class="ph codeph">ORDER BY</code> clause no longer requires an accompanying <code class="ph codeph">LIMIT</code> + clause in Impala 1.4.0 and higher, this query option is deprecated and has no effect. + </p> + + <p class="p"> + Prior to Impala 1.4.0, Impala queries that use the <code class="ph codeph"><a class="xref" href="impala_order_by.html#order_by">ORDER + BY</a></code> clause must also include a + <code class="ph codeph"><a class="xref" href="impala_limit.html#limit">LIMIT</a></code> clause, to avoid accidentally producing + huge result sets that must be sorted. Sorting a huge result set is a memory-intensive operation. In Impala + 1.4.0 and higher, Impala uses a temporary disk work area to perform the sort if that operation would + otherwise exceed the Impala memory limit on a particular host. + </p> + + <p class="p"> + <strong class="ph b">Type: numeric</strong> + </p> + + <p class="p"> + <strong class="ph b">Default:</strong> -1 (no default limit) + </p> + </div> +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_query_options.html">Query Options for the SET Statement</a></div></div></nav></article></main></body></html> \ No newline at end of file http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_delegation.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_delegation.html b/docs/build/html/topics/impala_delegation.html new file mode 100644 index 0000000..98fb545 --- /dev/null +++ b/docs/build/html/topics/impala_delegation.html @@ -0,0 +1,70 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_authentication.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="delegation"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>Configuring Impala Delegation for Hue and BI Tools</title></head><body id="delegation"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">Configuring Impala Delegation for Hue and BI Tools</h1> + + + + <div class="body conbody"> + + <p class="p"> + + When users submit Impala queries through a separate application, such as Hue or a business intelligence tool, + typically all requests are treated as coming from the same user. In Impala 1.2 and higher, authentication is + extended by a new feature that allows applications to pass along credentials for the users that connect to + them (known as <span class="q">"delegation"</span>), and issue Impala queries with the privileges for those users. Currently, + the delegation feature is available only for Impala queries submitted through application interfaces such as + Hue and BI tools; for example, Impala cannot issue queries using the privileges of the HDFS user. + </p> + + <p class="p"> + The delegation feature is enabled by a startup option for <span class="keyword cmdname">impalad</span>: + <code class="ph codeph">--authorized_proxy_user_config</code>. When you specify this option, users whose names you specify + (such as <code class="ph codeph">hue</code>) can delegate the execution of a query to another user. The query runs with the + privileges of the delegated user, not the original user such as <code class="ph codeph">hue</code>. The name of the + delegated user is passed using the HiveServer2 configuration property <code class="ph codeph">impala.doas.user</code>. + </p> + + <p class="p"> + You can specify a list of users that the application user can delegate to, or <code class="ph codeph">*</code> to allow a + superuser to delegate to any other user. For example: + </p> + +<pre class="pre codeblock"><code>impalad --authorized_proxy_user_config 'hue=user1,user2;admin=*' ...</code></pre> + + <div class="note note note_note"><span class="note__title notetitle">Note:</span> + Make sure to use single quotes or escape characters to ensure that any <code class="ph codeph">*</code> characters do not + undergo wildcard expansion when specified in command-line arguments. + </div> + + <p class="p"> + See <a class="xref" href="impala_config_options.html#config_options">Modifying Impala Startup Options</a> for details about adding or changing + <span class="keyword cmdname">impalad</span> startup options. See + <a class="xref" href="http://blog.cloudera.com/blog/2013/07/how-hiveserver2-brings-security-and-concurrency-to-apache-hive/"; target="_blank">this + blog post</a> for background information about the delegation capability in HiveServer2. + </p> + <p class="p"> + To set up authentication for the delegated users: + </p> + + <ul class="ul"> + <li class="li"> + <p class="p"> + On the server side, configure either user/password authentication through LDAP, or Kerberos + authentication, for all the delegated users. See <a class="xref" href="impala_ldap.html#ldap">Enabling LDAP Authentication for Impala</a> or + <a class="xref" href="impala_kerberos.html#kerberos">Enabling Kerberos Authentication for Impala</a> for details. + </p> + </li> + + <li class="li"> + <p class="p"> + On the client side, to learn how to enable delegation, consult the documentation + for the ODBC driver you are using. + </p> + </li> + </ul> + + </div> + +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_authentication.html">Impala Authentication</a></div></div></nav></article></main></body></html> \ No newline at end of file http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/75c46918/docs/build/html/topics/impala_delete.html ---------------------------------------------------------------------- diff --git a/docs/build/html/topics/impala_delete.html b/docs/build/html/topics/impala_delete.html new file mode 100644 index 0000000..3f95fa5 --- /dev/null +++ b/docs/build/html/topics/impala_delete.html @@ -0,0 +1,177 @@ +<!DOCTYPE html + SYSTEM "about:legacy-compat"> +<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="UTF-8"><meta name="copyright" content="(C) Copyright 2017"><meta name="DC.rights.owner" content="(C) Copyright 2017"><meta name="DC.Type" content="concept"><meta name="DC.Relation" scheme="URI" content="../topics/impala_langref_sql.html"><meta name="prodname" content="Impala"><meta name="prodname" content="Impala"><meta name="version" content="Impala 2.8.x"><meta name="version" content="Impala 2.8.x"><meta name="DC.Format" content="XHTML"><meta name="DC.Identifier" content="delete"><link rel="stylesheet" type="text/css" href="../commonltr.css"><title>DELETE Statement (Impala 2.8 or higher only)</title></head><body id="delete"><main role="main"><article role="article" aria-labelledby="ariaid-title1"> + + <h1 class="title topictitle1" id="ariaid-title1">DELETE Statement (<span class="keyword">Impala 2.8</span> or higher only)</h1> + + + + <div class="body conbody"> + + <p class="p"> + + Deletes an arbitrary number of rows from a Kudu table. + This statement only works for Impala tables that use the Kudu storage engine. + </p> + + <p class="p"> + <strong class="ph b">Syntax:</strong> + </p> + +<pre class="pre codeblock"><code> +DELETE [FROM] [<var class="keyword varname">database_name</var>.]<var class="keyword varname">table_name</var> [ WHERE <var class="keyword varname">where_conditions</var> ] + +DELETE <var class="keyword varname">table_ref</var> FROM [<var class="keyword varname">joined_table_refs</var>] [ WHERE <var class="keyword varname">where_conditions</var> ] +</code></pre> + + <p class="p"> + The first form evaluates rows from one table against an optional + <code class="ph codeph">WHERE</code> clause, and deletes all the rows that + match the <code class="ph codeph">WHERE</code> conditions, or all rows if + <code class="ph codeph">WHERE</code> is omitted. + </p> + + <p class="p"> + The second form evaluates one or more join clauses, and deletes + all matching rows from one of the tables. The join clauses can + include non-Kudu tables, but the table from which the rows + are deleted must be a Kudu table. The <code class="ph codeph">FROM</code> + keyword is required in this case, to separate the name of + the table whose rows are being deleted from the table names + of the join clauses. + </p> + + <p class="p"> + <strong class="ph b">Usage notes:</strong> + </p> + + <p class="p"> + The conditions in the <code class="ph codeph">WHERE</code> clause are the same ones allowed + for the <code class="ph codeph">SELECT</code> statement. See <a class="xref" href="impala_select.html#select">SELECT Statement</a> + for details. + </p> + + <p class="p"> + The conditions in the <code class="ph codeph">WHERE</code> clause can refer to + any combination of primary key columns or other columns. Referring to + primary key columns in the <code class="ph codeph">WHERE</code> clause is more efficient + than referring to non-primary key columns. + </p> + + <p class="p"> + If the <code class="ph codeph">WHERE</code> clause is omitted, all rows are removed from the table. + </p> + + <p class="p"> + Because Kudu currently does not enforce strong consistency during concurrent DML operations, + be aware that the results after this statement finishes might be different than you + intuitively expect: + </p> + <ul class="ul"> + <li class="li"> + <p class="p"> + If some rows cannot be deleted because their + some primary key columns are not found, due to their being deleted + by a concurrent <code class="ph codeph">DELETE</code> operation, + the statement succeeds but returns a warning. + </p> + </li> + <li class="li"> + <p class="p"> + A <code class="ph codeph">DELETE</code> statement might also overlap with + <code class="ph codeph">INSERT</code>, <code class="ph codeph">UPDATE</code>, + or <code class="ph codeph">UPSERT</code> statements running concurrently on the same table. + After the statement finishes, there might be more or fewer rows than expected in the table + because it is undefined whether the <code class="ph codeph">DELETE</code> applies to rows that are + inserted or updated while the <code class="ph codeph">DELETE</code> is in progress. + </p> + </li> + </ul> + + <p class="p"> + The number of affected rows is reported in an <span class="keyword cmdname">impala-shell</span> message + and in the query profile. + </p> + + <p class="p"> + <strong class="ph b">Statement type:</strong> DML + </p> + + <div class="note important note_important"><span class="note__title importanttitle">Important:</span> + After adding or replacing data in a table used in performance-critical queries, issue a <code class="ph codeph">COMPUTE + STATS</code> statement to make sure all statistics are up-to-date. Consider updating statistics for a + table after any <code class="ph codeph">INSERT</code>, <code class="ph codeph">LOAD DATA</code>, or <code class="ph codeph">CREATE TABLE AS + SELECT</code> statement in Impala, or after loading data through Hive and doing a <code class="ph codeph">REFRESH + <var class="keyword varname">table_name</var></code> in Impala. This technique is especially important for tables that + are very large, used in join queries, or both. + </div> + + <p class="p"> + <strong class="ph b">Examples:</strong> + </p> + + <p class="p"> + The following examples show how to delete rows from a specified + table, either all rows or rows that match a <code class="ph codeph">WHERE</code> + clause: + </p> + +<pre class="pre codeblock"><code> +-- Deletes all rows. The FROM keyword is optional. +DELETE FROM kudu_table; +DELETE kudu_table; + +-- Deletes 0, 1, or more rows. +-- (If c1 is a single-column primary key, the statement could only +-- delete 0 or 1 rows.) +DELETE FROM kudu_table WHERE c1 = 100; + +-- Deletes all rows that match all the WHERE conditions. +DELETE FROM kudu_table WHERE + (c1 > c2 OR c3 IN ('hello','world')) AND c4 IS NOT NULL; +DELETE FROM t1 WHERE + (c1 IN (1,2,3) AND c2 > c3) OR c4 IS NOT NULL; +DELETE FROM time_series WHERE + year = 2016 AND month IN (11,12) AND day > 15; + +-- WHERE condition with a subquery. +DELETE FROM t1 WHERE + c5 IN (SELECT DISTINCT other_col FROM other_table); + +-- Does not delete any rows, because the WHERE condition is always false. +DELETE FROM kudu_table WHERE 1 = 0; +</code></pre> + + <p class="p"> + The following examples show how to delete rows that are part + of the result set from a join: + </p> + +<pre class="pre codeblock"><code> +-- Remove _all_ rows from t1 that have a matching X value in t2. +DELETE t1 FROM t1 JOIN t2 ON t1.x = t2.x; + +-- Remove _some_ rows from t1 that have a matching X value in t2. +DELETE t1 FROM t1 JOIN t2 ON t1.x = t2.x + WHERE t1.y = FALSE and t2.z > 100; + +-- Delete from a Kudu table based on a join with a non-Kudu table. +DELETE t1 FROM kudu_table t1 JOIN non_kudu_table t2 ON t1.x = t2.x; + +-- The tables can be joined in any order as long as the Kudu table +-- is specified as the deletion target. +DELETE t2 FROM non_kudu_table t1 JOIN kudu_table t2 ON t1.x = t2.x; +</code></pre> + + <p class="p"> + <strong class="ph b">Related information:</strong> + </p> + + <p class="p"> + <a class="xref" href="impala_kudu.html#impala_kudu">Using Impala to Query Kudu Tables</a>, <a class="xref" href="impala_insert.html#insert">INSERT Statement</a>, + <a class="xref" href="impala_update.html#update">UPDATE Statement (Impala 2.8 or higher only)</a>, <a class="xref" href="impala_upsert.html#upsert">UPSERT Statement (Impala 2.8 or higher only)</a> + </p> + + </div> + +<nav role="navigation" class="related-links"><div class="familylinks"><div class="parentlink"><strong>Parent topic:</strong> <a class="link" href="../topics/impala_langref_sql.html">Impala SQL Statements</a></div></div></nav></article></main></body></html> \ No newline at end of file
