mirror of
https://github.com/sbrl/Pepperminty-Wiki.git
synced 2024-11-25 05:22:59 +00:00
More work on inverted indexes. I think we're getting there.
This commit is contained in:
parent
777a27064e
commit
29acc3a5e2
4 changed files with 319 additions and 103 deletions
212
build/index.php
212
build/index.php
|
|
@ -657,7 +657,71 @@ else
|
||||||
header("x-pageindex-decode-time: " . round(microtime(true) - $pageindex_read_start, 6) . "ms");
|
header("x-pageindex-decode-time: " . round(microtime(true) - $pageindex_read_start, 6) . "ms");
|
||||||
}
|
}
|
||||||
|
|
||||||
// Work around an Opera + Syntaxtic bug where there is no margin at the left hand side if there isn't a query string when accessing a .php file
|
//////////////////////////
|
||||||
|
///// Page id system /////
|
||||||
|
//////////////////////////
|
||||||
|
if(!file_exists("idindex.json"))
|
||||||
|
file_put_contents("idindex.json", "{}");
|
||||||
|
$idindex = json_decode(file_get_contents("idindex.json"));
|
||||||
|
class ids
|
||||||
|
{
|
||||||
|
/*
|
||||||
|
* @summary Gets the page id associated with the given pagename.
|
||||||
|
*/
|
||||||
|
public static function getid($pagename)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
foreach ($idindex as $id => $entry)
|
||||||
|
{
|
||||||
|
if($entry == $pagename)
|
||||||
|
return $id;
|
||||||
|
}
|
||||||
|
|
||||||
|
// This pagename doesn't have an id - assign it one quick!
|
||||||
|
return self::assign($pagename);
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Gets the page name associated with the given page id.
|
||||||
|
*/
|
||||||
|
public static function getpagename($id)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
if(!isset($idindex->$id))
|
||||||
|
return false;
|
||||||
|
else
|
||||||
|
return $idindex->$id;
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Assigns an id to a pagename. Doesn't check to make sure that
|
||||||
|
* pagename doesn't exist in the pageindex.
|
||||||
|
*/
|
||||||
|
protected static function assign($pagename)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
$nextid = count(array_keys(get_object_vars($idindex)));
|
||||||
|
|
||||||
|
if(isset($idindex->$nextid))
|
||||||
|
throw new Exception("The pageid is corrupt! Pepperminty Wiki generated the id $nextid, but that id is already in use.");
|
||||||
|
|
||||||
|
// Update the id index
|
||||||
|
$idindex->$nextid = utf8_encode($pagename);
|
||||||
|
|
||||||
|
// Save the id index
|
||||||
|
file_put_contents("idindex.json", json_encode($idindex));
|
||||||
|
|
||||||
|
return $nextid;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
//////////////////////////
|
||||||
|
//////////////////////////
|
||||||
|
|
||||||
|
// Work around an Opera + Syntaxtic bug where there is no margin at the left
|
||||||
|
// hand side if there isn't a query string when accessing a .php file.
|
||||||
if(!isset($_GET["action"]) and !isset($_GET["page"]))
|
if(!isset($_GET["action"]) and !isset($_GET["page"]))
|
||||||
{
|
{
|
||||||
http_response_code(302);
|
http_response_code(302);
|
||||||
|
|
@ -1277,47 +1341,59 @@ register_module([
|
||||||
|
|
||||||
class search
|
class search
|
||||||
{
|
{
|
||||||
|
// Words that we should exclude from the inverted index.
|
||||||
public static $stop_words = [
|
public static $stop_words = [
|
||||||
"a", "about", "above", "above", "across", "after", "afterwards", "again", "against",
|
"a", "about", "above", "above", "across", "after", "afterwards", "again",
|
||||||
"all", "almost", "alone", "along", "already", "also", "although", "always", "am", "among",
|
"against", "all", "almost", "alone", "along", "already", "also",
|
||||||
"amongst", "amoungst", "amount", "an", "and", "another", "any", "anyhow", "anyone", "anything",
|
"although", "always", "am", "among", "amongst", "amoungst", "amount",
|
||||||
"anyway", "anywhere", "are", "around", "as", "at", "back", "be", "became", "because", "become",
|
"an", "and", "another", "any", "anyhow", "anyone", "anything", "anyway",
|
||||||
"becomes", "becoming", "been", "before", "beforehand", "behind", "being", "below", "beside",
|
"anywhere", "are", "around", "as", "at", "back", "be", "became",
|
||||||
"besides", "between", "beyond", "bill", "both", "bottom", "but", "by", "call", "can", "cannot",
|
"because", "become", "becomes", "becoming", "been", "before",
|
||||||
"cant", "co", "con", "could", "couldnt", "cry", "de", "describe", "detail", "do", "done", "down",
|
"beforehand", "behind", "being", "below", "beside", "besides",
|
||||||
"due", "during", "each", "eg", "eight", "either", "eleven", "else", "elsewhere", "empty",
|
"between", "beyond", "bill", "both", "bottom", "but", "by", "call",
|
||||||
"enough", "etc", "even", "ever", "every", "everyone", "everything", "everywhere", "except",
|
"can", "cannot", "cant", "co", "con", "could", "couldnt", "cry", "de",
|
||||||
"few", "fifteen", "fify", "fill", "find", "fire", "first", "five", "for", "former", "formerly",
|
"describe", "detail", "do", "done", "down", "due", "during", "each",
|
||||||
"forty", "found", "four", "from", "front", "full", "further", "get", "give", "go", "had", "has",
|
"eg", "eight", "either", "eleven", "else", "elsewhere", "empty",
|
||||||
"hasnt", "have", "he", "hence", "her", "here", "hereafter", "hereby", "herein", "hereupon",
|
"enough", "etc", "even", "ever", "every", "everyone", "everything",
|
||||||
"hers", "herself", "him", "himself", "his", "how", "however", "hundred", "ie", "if", "in",
|
"everywhere", "except", "few", "fifteen", "fify", "fill", "find",
|
||||||
"inc", "indeed", "interest", "into", "is", "it", "its", "itself", "keep", "last", "latter",
|
"fire", "first", "five", "for", "former", "formerly", "forty", "found",
|
||||||
"latterly", "least", "less", "ltd", "made", "many", "may", "me", "meanwhile", "might", "mill",
|
"four", "from", "front", "full", "further", "get", "give", "go", "had",
|
||||||
"mine", "more", "moreover", "most", "mostly", "move", "much", "must", "my", "myself", "name",
|
"has", "hasnt", "have", "he", "hence", "her", "here", "hereafter",
|
||||||
"namely", "neither", "never", "nevertheless", "next", "nine", "no", "nobody", "none", "noone",
|
"hereby", "herein", "hereupon", "hers", "herself", "him", "himself",
|
||||||
"nor", "not", "nothing", "now", "nowhere", "of", "off", "often", "on", "once", "one", "only",
|
"his", "how", "however", "hundred", "ie", "if", "in", "inc", "indeed",
|
||||||
"onto", "or", "other", "others", "otherwise", "our", "ours", "ourselves", "out", "over", "own",
|
"interest", "into", "is", "it", "its", "itself", "keep", "last",
|
||||||
"part", "per", "perhaps", "please", "put", "rather", "re", "same", "see", "seem", "seemed",
|
"latter", "latterly", "least", "less", "ltd", "made", "many", "may",
|
||||||
"seeming", "seems", "serious", "several", "she", "should", "show", "side", "since", "sincere",
|
"me", "meanwhile", "might", "mine", "more", "moreover", "most",
|
||||||
"six", "sixty", "so", "some", "somehow", "someone", "something", "sometime", "sometimes",
|
"mostly", "move", "much", "must", "my", "myself", "name", "namely",
|
||||||
"somewhere", "still", "such", "system", "take", "ten", "than", "that", "the", "their", "them",
|
"neither", "never", "nevertheless", "next", "nine", "no", "none",
|
||||||
"themselves", "then", "thence", "there", "thereafter", "thereby", "therefore", "therein",
|
"nor", "not", "nothing", "now", "nowhere", "of", "off", "often", "on",
|
||||||
"thereupon", "these", "they", "thickv", "thin", "third", "this", "those", "though", "three",
|
"once", "one", "only", "onto", "or", "other", "others", "otherwise",
|
||||||
"through", "throughout", "thru", "thus", "to", "together", "too", "top", "toward", "towards",
|
"our", "ours", "ourselves", "out", "over", "own", "part", "per",
|
||||||
"twelve", "twenty", "two", "un", "under", "until", "up", "upon", "us", "very", "via", "was",
|
"perhaps", "please", "put", "rather", "re", "same", "see", "seem",
|
||||||
"we", "well", "were", "what", "whatever", "when", "whence", "whenever", "where", "whereafter",
|
"seemed", "seeming", "seems", "serious", "several", "she", "should",
|
||||||
"whereas", "whereby", "wherein", "whereupon", "wherever", "whether", "which", "while", "whither",
|
"show", "side", "since", "sincere", "six", "sixty", "so", "some",
|
||||||
"who", "whoever", "whole", "whom", "whose", "why", "will", "with", "within", "without", "would",
|
"somehow", "someone", "something", "sometime", "sometimes",
|
||||||
"yet", "you", "your", "yours", "yourself", "yourselves"
|
"somewhere", "still", "such", "system", "take", "ten", "than", "that",
|
||||||
|
"the", "their", "them", "themselves", "then", "thence", "there",
|
||||||
|
"thereafter", "thereby", "therefore", "therein", "thereupon", "these",
|
||||||
|
"they", "thickv", "thin", "third", "this", "those", "though", "three",
|
||||||
|
"through", "throughout", "thru", "thus", "to", "together", "too", "top",
|
||||||
|
"toward", "towards", "twelve", "twenty", "two", "un", "under", "until",
|
||||||
|
"up", "upon", "us", "very", "via", "was", "we", "well", "were", "what",
|
||||||
|
"whatever", "when", "whence", "whenever", "where", "whereafter",
|
||||||
|
"whereas", "whereby", "wherein", "whereupon", "wherever", "whether",
|
||||||
|
"which", "while", "whither", "who", "whoever", "whole", "whom", "whose",
|
||||||
|
"why", "will", "with", "within", "without", "would", "yet", "you",
|
||||||
|
"your", "yours", "yourself", "yourselves"
|
||||||
];
|
];
|
||||||
|
|
||||||
public static function index($source)
|
public static function index($source)
|
||||||
{
|
{
|
||||||
$source = html_entity_decode($source, ENT_QUOTES);
|
$source = html_entity_decode($source, ENT_QUOTES);
|
||||||
$source_length = strlen($source);
|
$source_length = strlen($source);
|
||||||
|
|
||||||
$index = [];
|
$index = [];
|
||||||
|
|
||||||
// Regex from
|
// Regex from
|
||||||
$terms = preg_split("/((^\p{P}+)|(\p{P}*\s+\p{P}*)|(\p{P}+$))/", $source, -1, PREG_SPLIT_NO_EMPTY);
|
$terms = preg_split("/((^\p{P}+)|(\p{P}*\s+\p{P}*)|(\p{P}+$))/", $source, -1, PREG_SPLIT_NO_EMPTY);
|
||||||
$i = 0;
|
$i = 0;
|
||||||
|
|
@ -1332,47 +1408,79 @@ class search
|
||||||
{
|
{
|
||||||
$index[$nterm] = [ "freq" => 0, "offsets" => [] ];
|
$index[$nterm] = [ "freq" => 0, "offsets" => [] ];
|
||||||
}
|
}
|
||||||
|
|
||||||
$index[$nterm]["freq"]++;
|
$index[$nterm]["freq"]++;
|
||||||
$index[$nterm]["offsets"][] = $i;
|
$index[$nterm]["offsets"][] = $i;
|
||||||
|
|
||||||
$i++;
|
$i++;
|
||||||
}
|
}
|
||||||
|
|
||||||
self::sort_index($index);
|
|
||||||
|
|
||||||
return $index;
|
return $index;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* @sumary Sorts an index alphabetically. This allows us to do a binary search instead of a regular sequential search.
|
* @summary Sorts an index alphabetically. Will also sort an inverted index.
|
||||||
|
* This allows us to do a binary search instead of a regular
|
||||||
|
* sequential search.
|
||||||
*/
|
*/
|
||||||
public static function sort_index(&$index)
|
public static function sort_index(&$index)
|
||||||
{
|
{
|
||||||
ksort($index, SORT_NATURAL);
|
ksort($index, SORT_NATURAL);
|
||||||
}
|
}
|
||||||
}
|
|
||||||
|
|
||||||
class ids
|
|
||||||
{
|
|
||||||
/*
|
/*
|
||||||
* @summary Gets the page id associated with the given pagename.
|
* @summary Compares two *regular* indexes to find the differences between them.
|
||||||
|
*
|
||||||
|
* @param {array} $indexa - The old index.
|
||||||
|
* @param {array} $indexb - The new index.
|
||||||
|
* @param {array} $changed - An array to be filled with the nterms of all
|
||||||
|
* the changed entries.
|
||||||
|
* @param {array} $removed - An array to be filled with the nterms of all
|
||||||
|
* the removed entries.
|
||||||
*/
|
*/
|
||||||
public static function get_page_id($pagename)
|
public static function compare_indexes($indexa, $indexb, &$changed, &$removed)
|
||||||
{
|
{
|
||||||
global $idindex;
|
foreach ($indexa as $nterm => $entrya)
|
||||||
|
{
|
||||||
|
if(!isset($indexb[$nterm]))
|
||||||
|
$removed[] = $nterm;
|
||||||
|
$entryb = $indexb[$nterm];
|
||||||
|
if($entrya !== $entryb) $changed[] = $nterm;
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* @summary Gets the page name associated with the given page id.
|
* @summary Reads in and parses an inverted index.
|
||||||
*/
|
*/
|
||||||
public static function get_id_pagename($id)
|
// Todo remove this function and make everything streamable
|
||||||
|
public static function parse_invindex($invindex_filename) {
|
||||||
|
$invindex = json_decode(file_get_contents($invindex_filename), true);
|
||||||
|
return $invindex;
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Merge an index into an inverted index.
|
||||||
|
*/
|
||||||
|
public static function merge_into_invindex(&$invindex, $pageid, &$index, &$removals = [])
|
||||||
{
|
{
|
||||||
global $idindex;
|
// Remove all the subentries that were removed since last time
|
||||||
|
foreach($removals as $nterm)
|
||||||
|
{
|
||||||
|
unset($invindex[$nterm][$pageid]);
|
||||||
|
}
|
||||||
|
|
||||||
|
// Merge all the new / changed index entries into the inverted index
|
||||||
|
foreach($index as $nterm => $newentry)
|
||||||
|
{
|
||||||
|
// If the nterm isn't in the inverted index, then create a space for it
|
||||||
|
if(!isset($invindex[$nterm])) $invindex[$nterm] = [];
|
||||||
|
$invindex[$nterm][$pageid] = $newentry;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static function save_invindex($filename, &$invindex)
|
||||||
|
{
|
||||||
|
file_put_contents($filename, json_encode($invindex));
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
||||||
66
core.php
66
core.php
|
|
@ -348,7 +348,71 @@ else
|
||||||
header("x-pageindex-decode-time: " . round(microtime(true) - $pageindex_read_start, 6) . "ms");
|
header("x-pageindex-decode-time: " . round(microtime(true) - $pageindex_read_start, 6) . "ms");
|
||||||
}
|
}
|
||||||
|
|
||||||
// Work around an Opera + Syntaxtic bug where there is no margin at the left hand side if there isn't a query string when accessing a .php file
|
//////////////////////////
|
||||||
|
///// Page id system /////
|
||||||
|
//////////////////////////
|
||||||
|
if(!file_exists("idindex.json"))
|
||||||
|
file_put_contents("idindex.json", "{}");
|
||||||
|
$idindex = json_decode(file_get_contents("idindex.json"));
|
||||||
|
class ids
|
||||||
|
{
|
||||||
|
/*
|
||||||
|
* @summary Gets the page id associated with the given pagename.
|
||||||
|
*/
|
||||||
|
public static function getid($pagename)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
foreach ($idindex as $id => $entry)
|
||||||
|
{
|
||||||
|
if($entry == $pagename)
|
||||||
|
return $id;
|
||||||
|
}
|
||||||
|
|
||||||
|
// This pagename doesn't have an id - assign it one quick!
|
||||||
|
return self::assign($pagename);
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Gets the page name associated with the given page id.
|
||||||
|
*/
|
||||||
|
public static function getpagename($id)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
if(!isset($idindex->$id))
|
||||||
|
return false;
|
||||||
|
else
|
||||||
|
return $idindex->$id;
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Assigns an id to a pagename. Doesn't check to make sure that
|
||||||
|
* pagename doesn't exist in the pageindex.
|
||||||
|
*/
|
||||||
|
protected static function assign($pagename)
|
||||||
|
{
|
||||||
|
global $idindex;
|
||||||
|
|
||||||
|
$nextid = count(array_keys(get_object_vars($idindex)));
|
||||||
|
|
||||||
|
if(isset($idindex->$nextid))
|
||||||
|
throw new Exception("The pageid is corrupt! Pepperminty Wiki generated the id $nextid, but that id is already in use.");
|
||||||
|
|
||||||
|
// Update the id index
|
||||||
|
$idindex->$nextid = utf8_encode($pagename);
|
||||||
|
|
||||||
|
// Save the id index
|
||||||
|
file_put_contents("idindex.json", json_encode($idindex));
|
||||||
|
|
||||||
|
return $nextid;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
//////////////////////////
|
||||||
|
//////////////////////////
|
||||||
|
|
||||||
|
// Work around an Opera + Syntaxtic bug where there is no margin at the left
|
||||||
|
// hand side if there isn't a query string when accessing a .php file.
|
||||||
if(!isset($_GET["action"]) and !isset($_GET["page"]))
|
if(!isset($_GET["action"]) and !isset($_GET["page"]))
|
||||||
{
|
{
|
||||||
http_response_code(302);
|
http_response_code(302);
|
||||||
|
|
|
||||||
|
|
@ -50,7 +50,7 @@
|
||||||
"author": "Starbeamrainbowlabs",
|
"author": "Starbeamrainbowlabs",
|
||||||
"description": "Adds proper search functionality to Pepperminty Wiki. Note that this module, at the moment, just contains test code while I figure out how best to write a search engine.",
|
"description": "Adds proper search functionality to Pepperminty Wiki. Note that this module, at the moment, just contains test code while I figure out how best to write a search engine.",
|
||||||
"id": "feature-search",
|
"id": "feature-search",
|
||||||
"lastupdate": 1446032086,
|
"lastupdate": 1446042661,
|
||||||
"optional": false
|
"optional": false
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
|
|
|
||||||
|
|
@ -27,47 +27,59 @@ register_module([
|
||||||
|
|
||||||
class search
|
class search
|
||||||
{
|
{
|
||||||
|
// Words that we should exclude from the inverted index.
|
||||||
public static $stop_words = [
|
public static $stop_words = [
|
||||||
"a", "about", "above", "above", "across", "after", "afterwards", "again", "against",
|
"a", "about", "above", "above", "across", "after", "afterwards", "again",
|
||||||
"all", "almost", "alone", "along", "already", "also", "although", "always", "am", "among",
|
"against", "all", "almost", "alone", "along", "already", "also",
|
||||||
"amongst", "amoungst", "amount", "an", "and", "another", "any", "anyhow", "anyone", "anything",
|
"although", "always", "am", "among", "amongst", "amoungst", "amount",
|
||||||
"anyway", "anywhere", "are", "around", "as", "at", "back", "be", "became", "because", "become",
|
"an", "and", "another", "any", "anyhow", "anyone", "anything", "anyway",
|
||||||
"becomes", "becoming", "been", "before", "beforehand", "behind", "being", "below", "beside",
|
"anywhere", "are", "around", "as", "at", "back", "be", "became",
|
||||||
"besides", "between", "beyond", "bill", "both", "bottom", "but", "by", "call", "can", "cannot",
|
"because", "become", "becomes", "becoming", "been", "before",
|
||||||
"cant", "co", "con", "could", "couldnt", "cry", "de", "describe", "detail", "do", "done", "down",
|
"beforehand", "behind", "being", "below", "beside", "besides",
|
||||||
"due", "during", "each", "eg", "eight", "either", "eleven", "else", "elsewhere", "empty",
|
"between", "beyond", "bill", "both", "bottom", "but", "by", "call",
|
||||||
"enough", "etc", "even", "ever", "every", "everyone", "everything", "everywhere", "except",
|
"can", "cannot", "cant", "co", "con", "could", "couldnt", "cry", "de",
|
||||||
"few", "fifteen", "fify", "fill", "find", "fire", "first", "five", "for", "former", "formerly",
|
"describe", "detail", "do", "done", "down", "due", "during", "each",
|
||||||
"forty", "found", "four", "from", "front", "full", "further", "get", "give", "go", "had", "has",
|
"eg", "eight", "either", "eleven", "else", "elsewhere", "empty",
|
||||||
"hasnt", "have", "he", "hence", "her", "here", "hereafter", "hereby", "herein", "hereupon",
|
"enough", "etc", "even", "ever", "every", "everyone", "everything",
|
||||||
"hers", "herself", "him", "himself", "his", "how", "however", "hundred", "ie", "if", "in",
|
"everywhere", "except", "few", "fifteen", "fify", "fill", "find",
|
||||||
"inc", "indeed", "interest", "into", "is", "it", "its", "itself", "keep", "last", "latter",
|
"fire", "first", "five", "for", "former", "formerly", "forty", "found",
|
||||||
"latterly", "least", "less", "ltd", "made", "many", "may", "me", "meanwhile", "might", "mill",
|
"four", "from", "front", "full", "further", "get", "give", "go", "had",
|
||||||
"mine", "more", "moreover", "most", "mostly", "move", "much", "must", "my", "myself", "name",
|
"has", "hasnt", "have", "he", "hence", "her", "here", "hereafter",
|
||||||
"namely", "neither", "never", "nevertheless", "next", "nine", "no", "nobody", "none", "noone",
|
"hereby", "herein", "hereupon", "hers", "herself", "him", "himself",
|
||||||
"nor", "not", "nothing", "now", "nowhere", "of", "off", "often", "on", "once", "one", "only",
|
"his", "how", "however", "hundred", "ie", "if", "in", "inc", "indeed",
|
||||||
"onto", "or", "other", "others", "otherwise", "our", "ours", "ourselves", "out", "over", "own",
|
"interest", "into", "is", "it", "its", "itself", "keep", "last",
|
||||||
"part", "per", "perhaps", "please", "put", "rather", "re", "same", "see", "seem", "seemed",
|
"latter", "latterly", "least", "less", "ltd", "made", "many", "may",
|
||||||
"seeming", "seems", "serious", "several", "she", "should", "show", "side", "since", "sincere",
|
"me", "meanwhile", "might", "mine", "more", "moreover", "most",
|
||||||
"six", "sixty", "so", "some", "somehow", "someone", "something", "sometime", "sometimes",
|
"mostly", "move", "much", "must", "my", "myself", "name", "namely",
|
||||||
"somewhere", "still", "such", "system", "take", "ten", "than", "that", "the", "their", "them",
|
"neither", "never", "nevertheless", "next", "nine", "no", "none",
|
||||||
"themselves", "then", "thence", "there", "thereafter", "thereby", "therefore", "therein",
|
"nor", "not", "nothing", "now", "nowhere", "of", "off", "often", "on",
|
||||||
"thereupon", "these", "they", "thickv", "thin", "third", "this", "those", "though", "three",
|
"once", "one", "only", "onto", "or", "other", "others", "otherwise",
|
||||||
"through", "throughout", "thru", "thus", "to", "together", "too", "top", "toward", "towards",
|
"our", "ours", "ourselves", "out", "over", "own", "part", "per",
|
||||||
"twelve", "twenty", "two", "un", "under", "until", "up", "upon", "us", "very", "via", "was",
|
"perhaps", "please", "put", "rather", "re", "same", "see", "seem",
|
||||||
"we", "well", "were", "what", "whatever", "when", "whence", "whenever", "where", "whereafter",
|
"seemed", "seeming", "seems", "serious", "several", "she", "should",
|
||||||
"whereas", "whereby", "wherein", "whereupon", "wherever", "whether", "which", "while", "whither",
|
"show", "side", "since", "sincere", "six", "sixty", "so", "some",
|
||||||
"who", "whoever", "whole", "whom", "whose", "why", "will", "with", "within", "without", "would",
|
"somehow", "someone", "something", "sometime", "sometimes",
|
||||||
"yet", "you", "your", "yours", "yourself", "yourselves"
|
"somewhere", "still", "such", "system", "take", "ten", "than", "that",
|
||||||
|
"the", "their", "them", "themselves", "then", "thence", "there",
|
||||||
|
"thereafter", "thereby", "therefore", "therein", "thereupon", "these",
|
||||||
|
"they", "thickv", "thin", "third", "this", "those", "though", "three",
|
||||||
|
"through", "throughout", "thru", "thus", "to", "together", "too", "top",
|
||||||
|
"toward", "towards", "twelve", "twenty", "two", "un", "under", "until",
|
||||||
|
"up", "upon", "us", "very", "via", "was", "we", "well", "were", "what",
|
||||||
|
"whatever", "when", "whence", "whenever", "where", "whereafter",
|
||||||
|
"whereas", "whereby", "wherein", "whereupon", "wherever", "whether",
|
||||||
|
"which", "while", "whither", "who", "whoever", "whole", "whom", "whose",
|
||||||
|
"why", "will", "with", "within", "without", "would", "yet", "you",
|
||||||
|
"your", "yours", "yourself", "yourselves"
|
||||||
];
|
];
|
||||||
|
|
||||||
public static function index($source)
|
public static function index($source)
|
||||||
{
|
{
|
||||||
$source = html_entity_decode($source, ENT_QUOTES);
|
$source = html_entity_decode($source, ENT_QUOTES);
|
||||||
$source_length = strlen($source);
|
$source_length = strlen($source);
|
||||||
|
|
||||||
$index = [];
|
$index = [];
|
||||||
|
|
||||||
// Regex from
|
// Regex from
|
||||||
$terms = preg_split("/((^\p{P}+)|(\p{P}*\s+\p{P}*)|(\p{P}+$))/", $source, -1, PREG_SPLIT_NO_EMPTY);
|
$terms = preg_split("/((^\p{P}+)|(\p{P}*\s+\p{P}*)|(\p{P}+$))/", $source, -1, PREG_SPLIT_NO_EMPTY);
|
||||||
$i = 0;
|
$i = 0;
|
||||||
|
|
@ -89,40 +101,72 @@ class search
|
||||||
$i++;
|
$i++;
|
||||||
}
|
}
|
||||||
|
|
||||||
self::sort_index($index);
|
|
||||||
|
|
||||||
return $index;
|
return $index;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* @sumary Sorts an index alphabetically. This allows us to do a binary search instead of a regular sequential search.
|
* @summary Sorts an index alphabetically. Will also sort an inverted index.
|
||||||
|
* This allows us to do a binary search instead of a regular
|
||||||
|
* sequential search.
|
||||||
*/
|
*/
|
||||||
public static function sort_index(&$index)
|
public static function sort_index(&$index)
|
||||||
{
|
{
|
||||||
ksort($index, SORT_NATURAL);
|
ksort($index, SORT_NATURAL);
|
||||||
}
|
}
|
||||||
}
|
|
||||||
|
|
||||||
class ids
|
|
||||||
{
|
|
||||||
/*
|
/*
|
||||||
* @summary Gets the page id associated with the given pagename.
|
* @summary Compares two *regular* indexes to find the differences between them.
|
||||||
|
*
|
||||||
|
* @param {array} $indexa - The old index.
|
||||||
|
* @param {array} $indexb - The new index.
|
||||||
|
* @param {array} $changed - An array to be filled with the nterms of all
|
||||||
|
* the changed entries.
|
||||||
|
* @param {array} $removed - An array to be filled with the nterms of all
|
||||||
|
* the removed entries.
|
||||||
*/
|
*/
|
||||||
public static function get_page_id($pagename)
|
public static function compare_indexes($indexa, $indexb, &$changed, &$removed)
|
||||||
{
|
{
|
||||||
global $idindex;
|
foreach ($indexa as $nterm => $entrya)
|
||||||
|
{
|
||||||
|
if(!isset($indexb[$nterm]))
|
||||||
|
$removed[] = $nterm;
|
||||||
|
$entryb = $indexb[$nterm];
|
||||||
|
if($entrya !== $entryb) $changed[] = $nterm;
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* @summary Gets the page name associated with the given page id.
|
* @summary Reads in and parses an inverted index.
|
||||||
*/
|
*/
|
||||||
public static function get_id_pagename($id)
|
// Todo remove this function and make everything streamable
|
||||||
|
public static function parse_invindex($invindex_filename) {
|
||||||
|
$invindex = json_decode(file_get_contents($invindex_filename), true);
|
||||||
|
return $invindex;
|
||||||
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* @summary Merge an index into an inverted index.
|
||||||
|
*/
|
||||||
|
public static function merge_into_invindex(&$invindex, $pageid, &$index, &$removals = [])
|
||||||
{
|
{
|
||||||
global $idindex;
|
// Remove all the subentries that were removed since last time
|
||||||
|
foreach($removals as $nterm)
|
||||||
|
{
|
||||||
|
unset($invindex[$nterm][$pageid]);
|
||||||
|
}
|
||||||
|
|
||||||
|
// Merge all the new / changed index entries into the inverted index
|
||||||
|
foreach($index as $nterm => $newentry)
|
||||||
|
{
|
||||||
|
// If the nterm isn't in the inverted index, then create a space for it
|
||||||
|
if(!isset($invindex[$nterm])) $invindex[$nterm] = [];
|
||||||
|
$invindex[$nterm][$pageid] = $newentry;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static function save_invindex($filename, &$invindex)
|
||||||
|
{
|
||||||
|
file_put_contents($filename, json_encode($invindex));
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Reference in a new issue