Hiong3-eng5 has uploaded a new change for review.
https://gerrit.wikimedia.org/r/79176
Change subject: OWD CSV
......................................................................
OWD CSV
Change-Id: I9d5b496d7bd09aecb49c3a8e2e41f785399ae42c
---
M OmegaWiki/Attribute.php
M OmegaWiki/DefinedMeaning.php
M OmegaWiki/Expression.php
M includes/formatCSV.php
A includes/jobs/OWOwdListJob.php
M includes/setup/OWJobs.php
M includes/specials/SpecialOWDownloads.php
7 files changed, 598 insertions(+), 32 deletions(-)
git pull
ssh://gerrit.wikimedia.org:29418/mediawiki/extensions/WikiLexicalData
refs/changes/76/79176/1
diff --git a/OmegaWiki/Attribute.php b/OmegaWiki/Attribute.php
index 8c4b661..91391c4 100644
--- a/OmegaWiki/Attribute.php
+++ b/OmegaWiki/Attribute.php
@@ -6,15 +6,15 @@
public $type = "";
/**
- * @param $id (String) or null if
- * @param $name (String)
- * @param $type (String or Structure)
+ * @param $id (String) or null if
+ * @param $name (String)
+ * @param $type (String or Structure)
* If String, can be "language", "spelling", "boolean",
- * "defined-meaning", "defining-expression", "relation-type",
"attribute",
+ * "defined-meaning", "defining-expression", "relation-type",
"attribute",
* "collection", "short-text", "text"
*
* If Structure, see below.
- */
+ */
public function __construct( $id, $name, $type ) {
$this->id = $id;
$this->name = $name;
@@ -29,7 +29,7 @@
$this->type = $type;
}
- // Since the attribute is a structure and unnamed, we use
+ // Since the attribute is a structure and unnamed, we use
// the default label associated with it.
if ( is_null( $this->id ) && ( $this->type instanceof Structure
) ) {
$this->id = $this->type->getStructureType();
@@ -38,7 +38,7 @@
$this->type->setStructureType( $this->id );
}
}
-
+
public function getId() {
return $this->id;
}
@@ -54,7 +54,7 @@
class Structure {
private $attributes;
private $type;
-
+
public function getAttributes() {
return $this->attributes;
}
@@ -80,7 +80,7 @@
* 'anonymous-structure' unless there is only a
* a single Attribute object, in which case the
structure
* will inherit its ID. Do not pass null.
- * @param $structure (Array or Parameter list) One or more Attribute
objects.
+ * @param $structure (Array or Parameter list) One or more Attribute
objects.
*
*/
public function __construct( $argumentList ) {
@@ -88,7 +88,7 @@
# We're trying to be clever.
$args = func_get_args();
$this->attributes = null;
-
+
if ( $args[0] instanceof Attribute ) {
$this->attributes = $args;
} elseif ( is_array( $args[0] ) ) {
@@ -125,34 +125,208 @@
public function supportsAttributeId( $attributeId ) {
// $result = false;
// $i = 0;
-//
+//
// while (!$result && $i < count($this->attributes)) {
// $result = $this->attributes[$i]->id == $attributeId;
// $i++;
// }
-//
+//
// return $result;
return true;
}
-
+
public function supportsAttribute( Attribute $attribute ) {
return $this->supportsAttributeId( $attribute->id );
}
-
+
public function __tostring() {
$result = "{";
if ( count( $this->attributes ) > 0 ) {
$result .= $this->attributes[0]->id;
-
+
for ( $i = 1; $i < count( $this->attributes ); $i++ )
$result .= ", " . $this->attributes[$i]->id;
}
$result .= "}";
-
+
return $result;
}
}
+class Attributes {
+ public function __construct() {
+ }
+
+ /**
+ * returns an array of the Text Attributes name and the text itself
+ * for a language
+ *
+ * else returns array()
+ *
+ * Note: $options can be used to introduce new variables
+ */
+ public static function getTextAttributes( $objectId, $options =
array(), $dc = null ) {
+ if ( is_null( $dc ) ) {
+ $dc = wdGetDataSetContext();
+ }
+ $dbr = wfGetDB( DB_SLAVE );
+
+ $cond = array();
+ if ( isset( $options['ORDER BY'] ) ) {
+ $cond['ORDER BY']= $options['ORDER BY'];
+ } else {
+ $cond['ORDER BY']= 'text';
+ }
+
+ if ( isset( $options['LIMIT'] ) ) {
+ $cond['LIMIT']= $options['LIMIT'];
+ }
+ if ( isset( $options['OFFSET'] ) ) {
+ $cond['OFFSET']= $options['OFFSET'];
+ }
+
+ $languageId = null;
+ if ( isset( $options['languageId'] ) ) {
+ $languageId = $options['languageId'];
+ }
+
+ $queryResult = $dbr->select(
+ "{$dc}_text_attribute_values",
+ array(
+ 'text',
+ 'attribute_mid',
+ ),
+ array(
+ 'object_id' => $objectId,
+ 'remove_transaction_id' => null
+ ),
+ __METHOD__,
+ $cond
+ );
+
+ $textAttributes = array();
+ $attributes = new Attributes;
+ foreach ( $queryResult as $ta ) {
+ $textAttributes[] = array(
+ 'text' => $ta->text,
+ 'attribute_name' =>
$attributes->getAttributeName( $ta->attribute_mid, $languageId )
+ );
+ }
+
+ if ( $textAttributes ) {
+ return $textAttributes;
+ }
+ return array();
+ }
+
+ /**
+ * returns the Attribute Name
+ *
+ * else returns null
+ *
+ */
+ public static function getAttributeName( $attributeId, $languageId =
null, $dc = null ) {
+ if ( is_null( $dc ) ) {
+ $dc = wdGetDataSetContext();
+ }
+ $dbr = wfGetDB( DB_SLAVE );
+
+ $vars = array(
+ 'synt.defined_meaning_id' => $attributeId,
+ 'synt.expression_id = exp.expression_id',
+ 'synt.remove_transaction_id' => null,
+ 'exp.remove_transaction_id' => null
+ );
+
+ if ( $languageId ) {
+ $vars['exp.language_id'] = $languageId;
+ }
+
+ $attributeName = $dbr->selectField(
+ array(
+ 'exp' => "{$dc}_expression",
+ 'synt' => "{$dc}_syntrans",
+ ),
+ 'spelling',
+ $vars,
+ __METHOD__
+ );
+
+ if ( $attributeName ) {
+ return $attributeName;
+ }
+ return null;
+ }
+
+ /**
+ * returns an array of the Options Attributes Values etc name and the
text itself
+ * for a language
+ *
+ * @param $objectId can be either syntransId or definedMeaningId
+ *
+ * else returns array()
+ *
+ * Note: $options can be used to introduce new variables
+ */
+ public static function getOptionAttributes( $objectId, $options =
array(), $dc = null ) {
+ if ( is_null( $dc ) ) {
+ $dc = wdGetDataSetContext();
+ }
+ $dbr = wfGetDB( DB_SLAVE );
+
+ $vars = array(
+ 'oav.object_id' => $objectId,
+ 'oav.option_id = oao.option_id',
+ 'ca.object_id = oao.attribute_id',
+ 'oav.remove_transaction_id' => null,
+ 'oao.remove_transaction_id' => null,
+ 'ca.remove_transaction_id' => null
+ );
+
+ $cond = array();
+ if ( isset( $options['LIMIT'] ) ) {
+ $cond['LIMIT'] = $options['LIMIT'];
+ }
+ if ( isset( $options['OFFSET'] ) ) {
+ $cond['OFFSET'] = $options['OFFSET'];
+ }
+
+ $languageId = null;
+ if ( isset( $options['languageId'] ) ) {
+ $languageId = $options['languageId'];
+ }
+
+ $queryResult = $dbr->select(
+ array(
+ 'oav' => "{$dc}_option_attribute_values",
+ 'oao' => "{$dc}_option_attribute_options",
+ 'ca' => "{$dc}_class_attributes"
+ ),
+ array(
+ 'attribute_mid',
+ 'option_mid'
+ ),
+ $vars,
+ __METHOD__,
+ $cond
+ );
+
+ $optionAttributes = array();
+ $attributes = new Attributes;
+ foreach ( $queryResult as $oa ) {
+ $optionAttributes[] = array(
+ 'attribute_name' =>
$attributes->getAttributeName( $oa->attribute_mid, $languageId ),
+ 'attribute_option_name' =>
$attributes->getAttributeName( $oa->option_mid, $languageId )
+ );
+ }
+
+ if ( $optionAttributes ) {
+ return $optionAttributes;
+ }
+ return array();
+ }
+
+}
diff --git a/OmegaWiki/DefinedMeaning.php b/OmegaWiki/DefinedMeaning.php
index dff52a9..ae8651b 100644
--- a/OmegaWiki/DefinedMeaning.php
+++ b/OmegaWiki/DefinedMeaning.php
@@ -42,7 +42,7 @@
$dmInfo["expression"] = null;
$match = $this->definedMeaningModel->checkExistence(
true, true );
}
-
+
// The defining expression is either bad or missing. Let's
redirect
// to the correct URL.
if ( empty( $dmInfo["expression"] ) && !is_null( $match ) ) {
@@ -111,7 +111,7 @@
$wgOut->addHTML( wfMessage(
"ow_db_consistency__not_found" )->text() . " ID:$definedMeaningId" );
return;
}
-
+
$wgOut->addHTML(
getDefinedMeaningEditor( $this->viewInformation )->edit(
$this->getIdStack( $dmModel->getId() ),
@@ -120,7 +120,7 @@
);
$this->outputEditFooter();
}
-
+
public function history() {
global $wgOut, $wgTitle ;
@@ -167,7 +167,7 @@
$this->getIdStack( $definedMeaningId ),
$dmModel->getRecord()
);
-
+
}
public function getTitle() {
@@ -182,10 +182,10 @@
$definedMeaningIdStructure = new Structure(
$o->definedMeaningId );
$definedMeaningIdRecord = new ArrayRecord(
$definedMeaningIdStructure, $definedMeaningIdStructure );
$definedMeaningIdRecord->definedMeaningId = $definedMeaningId;
-
+
$idStack = new IdStack( WLD_DEFINED_MEANING );
$idStack->pushKey( $definedMeaningIdRecord );
-
+
return $idStack;
}
@@ -202,13 +202,13 @@
return $this->getDefinedMeaningIdFromTitle( $wgTitle->getText()
);
}
- /**
+ /**
* Creates sidebar HTML for indicating concepts which exist
* in multiple datasets, and providing a link to add new
* mappings.
*
* Potential refactor candidate!
- */
+ */
protected function getConceptPanel() {
global $wgTitle, $wgUser, $wdShowCopyPanel;
$active = true; # wrong place, but hey
@@ -226,7 +226,7 @@
$active = ( $dataset->getPrefix() ==
$dc->getPrefix() );
$name = $dataset->fetchName();
$prefix = $dataset->getPrefix();
-
+
$class = $active ? 'dataset-panel-active' :
'dataset-panel-inactive';
$slot = $active ? "$name" : $sk->makeLinkObj(
$dm->getTitleObject(), $name, "dataset=$prefix" );
$html .= "<tr><td
class=\"$class\">$slot</td></tr>";
@@ -246,15 +246,15 @@
$html .= "</div>\n";
return $html;
}
-
+
/** @returns user interface html for copying Defined Meanings
* between datasets. returns an empty string if the user
* actually doesn't have permission to edit.
*/
protected function getCopyPanel() {
- # mostly same code as in SpecialAddCollection... possibly might
- # make a nice separate function
+ # mostly same code as in SpecialAddCollection... possibly might
+ # make a nice separate function
global
$wgUser;
@@ -269,7 +269,7 @@
}
/* Deprecated for now
-
+
$html= getOptionPanel( array (
'Copy to' => getSelect('CopyTo', $datasetarray)
));
@@ -277,7 +277,7 @@
$html = $this->getCopyPanel2();
return $html;
}
-
+
/** links to futt bugly alternate copy mechanism, the
* latter being something that actually is somewhat
* understandable (though not yet refactored into
@@ -286,7 +286,7 @@
protected function getCopyPanel2() {
global
$wgScriptPath, $wgCommunity_dc;
-
+
$html = "Copy to:<br />\n";
$datasets = wdGetDatasets();
$dataset = $datasets[$wgCommunity_dc];
@@ -301,3 +301,64 @@
}
+class DefinedMeanings {
+
+ public function __construct() {
+ }
+
+ /**
+ * returns an array of "Defined Meaning Id" objects
+ * for a language
+ *
+ * else returns null
+ */
+ public static function getLanguageIdDefinedMeaningId( $languageId,
$options = array(), $dc = null ) {
+ if ( is_null( $dc ) ) {
+ $dc = wdGetDataSetContext();
+ }
+ $dbr = wfGetDB( DB_SLAVE );
+
+ if ( isset( $options['ORDER BY'] ) ) {
+ $cond['ORDER BY']= $options['ORDER BY'];
+ } else {
+ $cond['ORDER BY']= 'defined_meaning_id';
+ }
+
+ if ( isset( $options['LIMIT'] ) ) {
+ $cond['LIMIT']= $options['LIMIT'];
+ }
+ if ( isset( $options['OFFSET'] ) ) {
+ $cond['OFFSET']= $options['OFFSET'];
+ }
+
+ $cond[] = 'DISTINCT';
+
+ $queryResult = $dbr->select(
+ array(
+ 'synt' => "{$dc}_syntrans",
+ 'exp' => "{$dc}_expression"
+ ),
+ 'defined_meaning_id',
+ array(
+ 'synt.expression_id = exp.expression_id',
+ 'language_id' => $languageId,
+ 'synt.remove_transaction_id' => null,
+ 'exp.remove_transaction_id' => null
+ ),
+ __METHOD__,
+ $cond
+ );
+
+
+ $definedMeaningId = array();
+ foreach ( $queryResult as $dm ) {
+ $definedMeaningId[] = $dm->defined_meaning_id;
+ }
+ $queryResult = array();
+
+ if ( $definedMeaningId ) {
+ return $definedMeaningId;
+ }
+ return null;
+ }
+}
diff --git a/OmegaWiki/Expression.php b/OmegaWiki/Expression.php
index 01a0fd8..6fe9a5c 100644
--- a/OmegaWiki/Expression.php
+++ b/OmegaWiki/Expression.php
@@ -75,6 +75,8 @@
$cond['OFFSET']= $options['OFFSET'];
}
+ $cond[] = 'DISTINCT';
+
$queryResult = $dbr->select(
"{$dc}_expression",
'spelling',
@@ -96,4 +98,60 @@
}
return null;
}
+
+ /**
+ * returns an array of "Expression" objects
+ * for a defined meaning id for a language
+ *
+ * else returns null
+ */
+ public static function getDefinedMeaningIdAndLanguageIdExpressions(
$languageId, $definedMeaningId, $options = array(), $dc = null ) {
+ if ( is_null( $dc ) ) {
+ $dc = wdGetDataSetContext();
+ }
+ $dbr = wfGetDB( DB_SLAVE );
+
+ if ( isset( $options['ORDER BY'] ) ) {
+ $cond['ORDER BY']= $options['ORDER BY'];
+ } else {
+ $cond['ORDER BY']= 'spelling';
+ }
+
+ if ( isset( $options['LIMIT'] ) ) {
+ $cond['LIMIT']= $options['LIMIT'];
+ }
+ if ( isset( $options['OFFSET'] ) ) {
+ $cond['OFFSET']= $options['OFFSET'];
+ }
+
+ $cond[] = 'DISTINCT';
+
+ $queryResult = $dbr->select(
+ array(
+ 'synt' => "{$dc}_syntrans",
+ 'exp' => "{$dc}_expression"
+ ),
+ 'spelling',
+ array(
+ 'synt.expression_id = exp.expression_id',
+ 'language_id' => $languageId,
+ 'defined_meaning_id' => $definedMeaningId,
+ 'synt.remove_transaction_id' => null,
+ 'exp.remove_transaction_id' => null
+ ),
+ __METHOD__,
+ $cond
+ );
+
+ $expression = array();
+ foreach ( $queryResult as $exp ) {
+ $expression[] = $exp->spelling;
+ }
+
+ if ( $expression ) {
+ return $expression;
+ }
+ return null;
+ }
+
}
diff --git a/includes/formatCSV.php b/includes/formatCSV.php
index f6e2771..9a61ce4 100644
--- a/includes/formatCSV.php
+++ b/includes/formatCSV.php
@@ -16,6 +16,10 @@
if( preg_match( '/\"/', $column ) ) {
$column = str_replace( '"', '""', $column );
}
- return '"' . $column . '"';
+ $column = '"' . $column . '"';
+ if ( $column == '""') {
+ return '';
+ }
+ return $column;
}
}
diff --git a/includes/jobs/OWOwdListJob.php b/includes/jobs/OWOwdListJob.php
new file mode 100644
index 0000000..59b7c56
--- /dev/null
+++ b/includes/jobs/OWOwdListJob.php
@@ -0,0 +1,239 @@
+<?php
+/**
+ * Create a list of OmegaWiki Developer data which can be used to generate
+ * Dictionaries.
+ *
+ * TODO
+ * Monolingual:
+ * Final download file would be compressed as owl_fra_csv.zip
+ *
+ */
+
+global $wgWldOwScriptPath, $wgWldIncludesScriptPath;
+require_once( $wgWldOwScriptPath . 'Attribute.php' );
+require_once( $wgWldOwScriptPath . 'DefinedMeaning.php' );
+require_once( $wgWldOwScriptPath . 'Expression.php' );
+require_once( $wgWldIncludesScriptPath . 'formatCSV.php' );
+
+Class CreateOwdListJob extends Job {
+
+ public function __construct( $title, $params ) {
+ parent::__construct( 'CreateOwdList', $title, $params );
+ }
+
+ /**
+ * Execute the job
+ * @return bool
+ */
+ public function run() {
+ // Load data from $this->params and $this->title
+ if ( isset( $this->params['langcode'] ) ) {
+ $languageId = $this->params['langcode'];
+ }
+
+ if ( isset( $this->params['type'] ) ) {
+ $type = $this->params['type'];
+ }
+
+ if ( isset( $this->params['format'] ) ) {
+ $format = $this->params['format'];
+ }
+
+ if ( isset( $this->params['start'] ) ) {
+ $start = $this->params['start'];
+ }
+
+ // Perform your updates
+
+ if ( $type && $languageId && $format && $start ) {
+ $this->createList( $type, $languageId, $format, $start
);
+ return true;
+ }
+ return false;
+
+ }
+
+ protected function createList( $type, $code, $format, $start ) {
+ global $wgWldDownloadScriptPath;
+ $csv = new WldFormatCSV();
+
+ // the greater the value of $sqlLimit the faster the download
file is
+ // finished but the slower each web page loads while the job is
being
+ // processed.
+ $sqlLimit = 100;
+ $ctrOver = $sqlLimit + 1;
+
+ $options['OFFSET'] = $start;
+
+ $options['LIMIT'] = $sqlLimit;
+
+ // Why order by defined_meaning_id? To avoid duplication of
words
+ // and skipping of some. When a language is constantly edited,
+ // order by spelling would not accurately get all unique
expressions
+ // when job queued.
+ $options['ORDER BY'] = 'defined_meaning_id';
+
+ // language specifics
+ $languageId = getLanguageIdForIso639_3( $code );
+ if ( !$languageId ) {
+ return false;
+ }
+ $this->DefinedMeanings = new DefinedMeanings;
+ $languageDefinedMeaningIds =
DefinedMeanings::getLanguageIdDefinedMeaningId( $languageId, $options );
+
+ // create File name
+ $zippedAs = $type . "_$code" . ".$format";
+ $synZippedAs = $type . '_syn_' . $code . ".$format";
+
+ $zipName = $wgWldDownloadScriptPath . $type . "_$code" .
"_$format" . ".zip";
+
+ // When someone updates the file while someone is
+ // downloading the file, the file may ( in my mind ),
+ // be corrupted. So process it first as a temporary file,
+ // delete the original file, and rename the temporary file ~he
+ $tempFileName = $wgWldDownloadScriptPath;
+ $tempFileName .= $format . "_$type" . "_$code.tmp";
+ $tempSynFileName = $wgWldDownloadScriptPath;
+ $tempSynFileName .= $format . "_$type" . '_syn_' . "$code.tmp";
+
+ if ( $start == 1 ) {
+ $translations = array();
+ $fh = fopen ( $tempFileName, 'w' );
+ $fhsyn = fopen ( $tempSynFileName, 'w' );
+ } else {
+ $fh = fopen ( $tempFileName, 'a' );
+ $fhsyn = fopen ( $tempSynFileName, 'a' );
+ }
+ // Add data
+ $ctr = 0;
+ if ( $start != 0 ) {
+
+ foreach ( $languageDefinedMeaningIds as
$definedMeaningId ) {
+
+ if ( $ctr != $ctrOver ) {
+ $text =
getDefinedMeaningDefinitionForLanguage( $definedMeaningId, $languageId );
+ $text = preg_replace( '/\n/', '\\n',
$text );
+ $text = $csv->formatCSVcolumn( $text );
+
+ $this->Expressions = new Expressions;
+ $expressions =
Expressions::getDefinedMeaningIdAndLanguageIdExpressions( $languageId,
$definedMeaningId );
+
+ $this->Attributes = new Attributes;
+ foreach ( $expressions as $spelling ) {
+ $IPA = null;
+ $hyphenation = null;
+ $example = null;
+ $usage = null;
+
+ $expressionId =
getExpressionId( $spelling, $languageId );
+ $syntransId = getSynonymId(
$definedMeaningId, $expressionId );
+ $textAttributes =
Attributes::getTextAttributes( $syntransId, array( 'languageId' =>
WLD_ENGLISH_LANG_ID ) );
+ foreach ( $textAttributes as
$row ) {
+ $row['text'] =
preg_replace( '/\n/', '\\n', $row['text'] );
+ $row['text'] =
preg_replace( '/\t/', '\\t', $row['text'] );
+ if (
$row['attribute_name'] == 'International Phonetic Alphabet') {
+ $IPA .=
$row['text'] . ';';
+ $row = array(
'text' => null, 'attribute_name' => null );
+ }
+ if (
$row['attribute_name'] == 'hyphenation') {
+ $hyphenation .=
$row['text'] . ';';
+ $row = array(
'text' => null, 'attribute_name' => null );
+ }
+ if (
$row['attribute_name'] == 'example sentence') {
+ $example .=
$row['text'] . ';';
+ $row = array(
'text' => null, 'attribute_name' => null );
+ }
+ if (
$row['attribute_name'] == 'usage') {
+ $usage .=
$row['text'] . ';';
+ $row = array(
'text' => null, 'attribute_name' => null );
+ }
+ // if ( $row['text'] !=
null ) {
+ // echo "TEXT: " .
$row['attribute_name'] . " => ";
+ // echo
$row['text'] . "\n";
+ // }
+ }
+ $textAttributes = array();
+ $IPA = preg_replace( '/;$/',
'', $IPA );
+ $hyphenation = preg_replace(
'/;$/', '', $hyphenation );
+ $example = preg_replace(
'/;$/', '', $example );
+ $usage = preg_replace( '/;$/',
'', $usage );
+
+ $POS = null;
+ $oUsage = null;
+ $optionAttributes =
Attributes::getOptionAttributes( $syntransId, array( 'languageId' =>
WLD_ENGLISH_LANG_ID ) );
+ foreach ( $optionAttributes as
$row ) {
+
$row['attribute_option_name'] = preg_replace( '/\n/', '\\n',
$row['attribute_option_name'] );
+
$row['attribute_option_name'] = preg_replace( '/\t/', '\\t',
$row['attribute_option_name'] );
+ if (
$row['attribute_name'] == 'part of speech' ) {
+ $POS .=
$row['attribute_option_name'] . ';';
+ $row = array(
'attribute_option_name' => null, 'attribute_name' => null );
+ }
+ if (
$row['attribute_name'] == 'usage' ) {
+ $oUsage .=
$row['attribute_option_name'] . ';';
+ $row = array(
'attribute_option_name' => null, 'attribute_name' => null );
+ }
+ // if (
$row['attribute_name'] != null ) {
+ // echo "OPTN: " .
$row['attribute_name'] . " => ";
+ // echo
$row['attribute_option_name'] . "\n";
+ // }
+ }
+ $optionAttributes = array();
+ $POS = preg_replace( '/;$/',
'', $POS );
+ $oUsage = preg_replace( '/;$/',
'', $oUsage );
+
+ fwrite( $fhsyn,
+ $definedMeaningId .
+ ',' . $languageId .
+ ',' . $syntransId .
+ ',' .
$csv->formatCSVcolumn( $spelling ) .
+ ',' .
$csv->formatCSVcolumn( $IPA ) .
+ ',' .
$csv->formatCSVcolumn( $hyphenation ) .
+ ',' .
$csv->formatCSVcolumn( $example ) .
+ ',' .
$csv->formatCSVcolumn( $usage ) .
+ ',' .
$csv->formatCSVcolumn( $POS ) .
+ ',' .
$csv->formatCSVcolumn( $oUsage ) .
+ "\n"
+ );
+
+ }
+
+ fwrite( $fh,
+ $definedMeaningId .
+ ',' . $languageId .
+ ',' . $text .
+ "\n"
+ );
+ }
+ $ctr ++;
+
+ }
+
+ }
+ fclose( $fh );
+ fclose( $fhsyn );
+
+ // incomplete job
+ if ( $ctr == $sqlLimit ) {
+ $jobParams = array( 'type' => $type, 'langcode' =>
$code, 'format' => $format );
+ $jobParams['start'] = $start + $sqlLimit;
+ $jobName = 'User:JobQuery/' . $type . '_' . $code . '.'
. $format;
+ $title = Title::newFromText( $jobName );
+ $job = new CreateOwdListJob( $title, $jobParams );
+ JobQueueGroup::singleton()->push( $job ); // mediawiki
>= 1.21
+ } else { // complete job
+ // Zip file
+ if ( file_exists( $zipName ) ) {
+ unlink( $zipName );
+ }
+ $zip = new ZipArchive();
+ $zip->open( $zipName, ZipArchive::CREATE );
+ $zip->addfile( $tempFileName, $zippedAs );
+ $zip->addfile( $tempSynFileName, $synZippedAs );
+ $zip->close();
+ unlink( $tempFileName );
+ unlink( $tempSynFileName );
+ }
+
+ }
+
+}
diff --git a/includes/setup/OWJobs.php b/includes/setup/OWJobs.php
index 4f1ff76..5422366 100644
--- a/includes/setup/OWJobs.php
+++ b/includes/setup/OWJobs.php
@@ -17,10 +17,14 @@
# Location of the job classes (Tell MediaWiki to load this file)
$wgAutoloadClasses['CreateExpressionListJob'] = $wgWldJobsScriptPath .
'OWExpressionListJob.php';
$wgAutoloadClasses['CreateDefinedExpressionListJob'] = $wgWldJobsScriptPath .
'OWDefinedExpressionListJob.php';
+$wgAutoloadClasses['CreateOwdListJob'] = $wgWldJobsScriptPath .
'OWOwdListJob.php';
+//$wgAutoloadClasses['CreateSynTransListJob'] = $wgWldJobsScriptPath .
'OWSynTransListJob.php';
# Tell MediaWiki about the jobs and its class name
$wgJobClasses['CreateExpressionList'] = 'CreateExpressionListJob';
$wgJobClasses['CreateDefinedExpressionList'] =
'CreateDefinedExpressionListJob';
+$wgJobClasses['CreateOwdList'] = 'CreateOwdListJob';
+//$wgJobClasses['CreateSynTransList'] = 'CreateSynTransListJob';
// Return true so that MediaWiki continues to load extensions.
return true;
diff --git a/includes/specials/SpecialOWDownloads.php
b/includes/specials/SpecialOWDownloads.php
index 709cd33..e790415 100644
--- a/includes/specials/SpecialOWDownloads.php
+++ b/includes/specials/SpecialOWDownloads.php
@@ -59,6 +59,18 @@
JobQueueGroup::singleton()->push( $job
); // mediawiki >= 1.21
}
}
+ // update Owd
+ if ( preg_match( '/^owd_/', $fileName ) ) {
+ preg_match( '/_(.+)\.(.+)/', $fileName, $match
);
+ $jobName = 'JobQuery/' . $fileName;
+ $jobParams = array( 'type' => 'owd', 'langcode'
=> $match[1], 'format' => $match[2], 'start' => '1' );
+ $jobExist = $wldJobs->downloadJobExist(
$jobName );
+ if ( $jobExist == false ) {
+ $title = Title::newFromText( 'User:' .
$jobName );
+ $job = new CreateOwdListJob( $title,
$jobParams );
+ JobQueueGroup::singleton()->push( $job
); // mediawiki >= 1.21
+ }
+ }
}
// Segregate by groups
@@ -74,6 +86,10 @@
// translations
if ( preg_match( '/^trans_/', $files ) ) {
$translations[] = "$files";
+ }
+ // OmegaWiki Development
+ if ( preg_match( '/^owd_.+zip$/', $files ) ) {
+ $development[] = "$files";
}
}
@@ -98,6 +114,13 @@
$wikitext[] = $myLine . "\n|}\n";
}
+ // Process Development
+ if ( $development ) {
+ $wikitext[] = '===Development===';
+ $myLine = $this->processText( $development );
+ $wikitext[] = $myLine . "\n|}\n";
+ }
+
// output wikitext
foreach( $wikitext as $wikilines ) {
$output->addWikiText( $wikilines );
@@ -116,6 +139,9 @@
if ( preg_match( '/^def_(.+)\./', $line, $match ) ) {
$language = $match[1];
}
+ if ( preg_match( '/^owd_(.+)_csv\./', $line, $match ) )
{
+ $language = $match[1];
+ }
$languageId = getLanguageIdForIso639_3( $language );
// TODO: How to internationalize $nameLanguageId?
$nameLanguageId = WLD_ENGLISH_LANG_ID;
--
To view, visit https://gerrit.wikimedia.org/r/79176
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I9d5b496d7bd09aecb49c3a8e2e41f785399ae42c
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/extensions/WikiLexicalData
Gerrit-Branch: master
Gerrit-Owner: Hiong3-eng5 <hiong3.e...@gmail.com>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
