$db = new Db(); $unreleasedSql = $gunreleased != '' ? ' and binaries.procstat NOT IN (4,5,6) and binaries.releaseID IS NULL' : ''; $resbin = $db->queryDirect(sprintf("SELECT binaries.ID as binID, binaries.name as binName from binaries where binaries.groupID = %d%s order by dateadded", $gselected, $unreleasedSql)); $matches = array(); while ($rowbin = mysql_fetch_assoc($resbin)) { if (preg_match($gregex, $rowbin["binName"], $binmatch)) { $binmatch = array_map("trim", $binmatch); if (isset($binmatch['reqid']) && ctype_digit($binmatch['reqid']) && (!isset($binmatch['name']) || empty($binmatch['name']))) { $binmatch['name'] = $binmatch['reqid']; } if (!isset($binmatch['name']) || empty($binmatch['name'])) { //echo "bad regex applied which didnt return right number of capture groups<br />"; } else { $binmatch['count'] = isset($matches[$binmatch['name']]['count']) ? $matches[$binmatch['name']]['count'] + 1 : 1; $binmatch['bininfo'] = $rowbin; $binmatch['catname'] = $catList[$cat->determineCategory($gname[$gselected], $binmatch['name'])]; $matches[$binmatch['name']] = $binmatch; } } } $offset = isset($_REQUEST["offset"]) ? $_REQUEST["offset"] : 0; $page->smarty->assign('pagertotalitems', sizeof($matches)); $page->smarty->assign('pageroffset', $offset); $page->smarty->assign('pageritemsperpage', ITEMS_PER_PAGE); $page->smarty->assign('pagerquerybase', WWW_TOP . "/regex-test.php?action=submit&groupname={$gselected}®ex=" . urlencode($gregex) . "&unreleased={$gunreleased}&offset="); $pager = $page->smarty->fetch($page->getCommonTemplate("pager.tpl")); $page->smarty->assign('pager', $pager); $matches = array_slice($matches, $offset, ITEMS_PER_PAGE); $page->smarty->assign('matches', $matches); } break;
$groupID = -1; foreach ($nzbInfo->groups as $group) { $group = (string) $group; if (array_key_exists($group, $siteGroups)) { $groupID = $siteGroups[$group]; $groupName = $group; break; } } if ($groupID == -1) { $retval .= "Missing Group for NZB : " . $browserpostednames[$nzbFile] . " - " . join(", ", $nzbInfo->groups) . "<br/>"; continue; } $relguid = md5(uniqid()); $name = $releases->cleanReleaseName(str_replace(".nzb", "", $browserpostednames[$nzbFile])); $catId = $cat->determineCategory($groupName, $name); $relid = $releases->insertRelease($name, $nzbInfo->filecount, $groupID, $relguid, $catId, "", date("Y-m-d H:i:s", $nzbInfo->postedlast), $nzbInfo->poster, "", $page->site); $db->exec(sprintf("update releases set totalpart = %d, size = %s, completion = %d, GID=%s where ID = %d", $nzbInfo->filecount, $nzbInfo->filesize, $nzbInfo->completion, $db->escapeString($nzbInfo->gid), $relid)); $nzbfilename = $nzb->getNZBPath($relguid, $page->site->nzbpath, true); $fp = _gzopen($nzbfilename, "w"); if ($fp) { gzwrite($fp, $nzbInfo->toNzb()); gzclose($fp); $retval .= sprintf(" Imported %s<br/>", $name); } } } } } $page->smarty->assign('output', $retval); }
public function testRegex($regex, $groupname, $poster, $ignorematched, $matchagainstbins) { $db = new Db(); $cat = new Category(); $catList = $cat->getForSelect(); $matches = array(); if ($groupname === 0) { $groupname = '.*'; } if ($matchagainstbins !== '') { $sql = sprintf("select b.*, '0' as size, '0' as blacklistID, g.name as groupname from binaries b left join groups g on g.ID = b.groupID where b.groupID IN (select g.ID from groups g where g.name REGEXP %s) order by b.date desc", $db->escapeString('^' . $groupname . '$')); } else { $sql = sprintf("select rrt.* from releaseregextesting rrt where rrt.groupname REGEXP %s order by rrt.date desc", $db->escapeString('^' . $groupname . '$')); } $resbin = $db->queryDirect($sql); while ($rowbin = $db->getAssocArray($resbin)) { if ($ignorematched !== '' && ($rowbin['regexID'] != '' || $rowbin['blacklistID'] == 1)) { continue; } $regexarr = array("ID" => "", 'regex' => $regex, 'poster' => $poster, "categoryID" => ""); $regexCheck = $this->performMatch($regexarr, $rowbin['name'], $rowbin['fromname']); if ($regexCheck !== false) { $relname = $regexCheck['name']; $relparts = explode("/", $regexCheck['parts']); $matches[$relname]['name'] = $relname; $matches[$relname]['parts'] = $regexCheck['parts']; $matches[$relname]['bincount'] = isset($matches[$relname]['bincount']) ? $matches[$relname]['bincount'] + 1 : 1; $matches[$relname]['bininfo'][] = $rowbin; $matches[$relname]['binsize'][] = $rowbin['size']; $matches[$relname]['totalsize'] = array_sum($matches[$relname]['binsize']); $matches[$relname]['relparts'][$relparts[1]] = $relparts[1]; $matches[$relname]['reltotalparts'] = array_sum($matches[$relname]['relparts']); $matches[$relname]['regexID'] = $regexCheck['regexID']; if (ctype_digit($regexCheck['regcatid'])) { $matches[$relname]['catname'] = $catList[$regexCheck['regcatid']]; } else { $matches[$relname]['catname'] = $catList[$cat->determineCategory($groupname, $relname)]; } } } //echo '<pre>'; //print_r(array_pop($matches)); //echo '</pre>'; return $matches; }
/** * Work out the category based on the name, resets to null if no category matched. */ private function determineCategory($rel, &$foundName, &$methodused) { $categoryID = null; $category = new Category(); $categoryID = $category->determineCategory($rel['groupname'], $foundName); if ($methodused == 'a.b.hdtv.x264' && $rel['groupname'] == 'alt.binaries.hdtv.x264') { $categoryID = Category::CAT_MOVIE_HD; } if ($categoryID == $rel['categoryID'] || $categoryID == '7900' || ($foundName == $rel['name'] || $foundName == $rel['searchname'])) { $foundName = null; $methodused = null; } else { $foundName = str_replace(''', '', trim(html_entity_decode($foundName))); $name = str_replace(' ', '_', $foundName); $searchname = str_replace('_', ' ', $foundName); if ($this->verbose) { echo 'ReleaseID: ' . $rel['RID'] . "\n"; echo ' Group: ' . $rel['groupname'] . "\n"; echo ' Old Name: ' . $rel['name'] . "\n"; echo ' Old SearchName: ' . $rel['searchname'] . "\n"; echo ' New Name: ' . $name . "\n"; echo ' New SearchName: ' . $searchname . "\n"; echo ' Old Cat: ' . $rel['categoryID'] . "\n"; echo ' New Cat: ' . $categoryID . "\n"; echo ' Method: ' . $methodused . "\n"; echo " Status: \t\tRelease changed\n\n"; } if (!$this->echoonly) { $db = new DB(); $db->exec(sprintf("update releases SET name = %s, searchname = %s, categoryID = %d, imdbID = NULL, tvinfoID = -1, bookinfoID = NULL, musicinfoID = NULL, consoleinfoID = NULL WHERE releases.ID = %d", $db->escapeString($name), $db->escapeString($searchname), $categoryID, $rel['RID'])); } $this->numupdated++; } }
function processReleases() { require_once WWW_DIR . "/lib/binaries.php"; $db = new DB(); $currTime_ori = $db->queryOneRow("SELECT NOW() as now"); $cat = new Category(); $nzb = new Nzb(); $s = new Sites(); $releaseRegex = new ReleaseRegex(); $page = new Page(); $groups = new Groups(); $retcount = 0; echo $s->getLicense(); echo "\n\nStarting release update process (" . date("Y-m-d H:i:s") . ")\n"; if (!file_exists($page->site->nzbpath)) { echo "Bad or missing nzb directory - " . $page->site->nzbpath; return -1; } // // Get all regexes for all groups which are to be applied to new binaries // in order of how they should be applied // $releaseRegex->get(); echo "Stage 1 : Applying regex to binaries\n"; $activeCategories = $cat->get(); $catbasedsizes = $db->getLookupAsArray($activeCategories, "ID"); $activeGroups = $groups->getActive(false); $groupbasedminsizes = $db->getLookupAsArray($groups->getAllNoReleases(), "ID"); foreach ($activeGroups as $groupArr) { //check if regexes have already been applied during update binaries if ($groupArr['regexmatchonly'] == 1) { continue; } $groupRegexes = $releaseRegex->getForGroup($groupArr['name']); echo "Stage 1 : Applying " . sizeof($groupRegexes) . " regexes to group " . $groupArr['name'] . "\n"; // Get out all binaries of STAGE0 for current group $newUnmatchedBinaries = array(); $ressql = sprintf("SELECT binaries.ID, binaries.name, binaries.date, binaries.totalParts, binaries.procstat, binaries.fromname from binaries where groupID = %d and procstat IN (%d,%d) and regexID IS NULL order by binaries.date asc", $groupArr['ID'], Releases::PROCSTAT_NEW, Releases::PROCSTAT_TITLENOTMATCHED); $resbin = $db->queryDirect($ressql); $matchedbins = 0; while ($rowbin = $db->getAssocArray($resbin)) { $regexMatches = array(); foreach ($groupRegexes as $groupRegex) { $regexCheck = $releaseRegex->performMatch($groupRegex, $rowbin['name']); if ($regexCheck !== false) { $regexMatches = $regexCheck; break; } } if (!empty($regexMatches)) { $matchedbins++; $relparts = explode("/", $regexMatches['parts']); $db->exec(sprintf("update binaries set relname = replace(%s, '_', ' '), relpart = %d, reltotalpart = %d, procstat=%d, categoryID=%s, regexID=%d, reqID=%s where ID = %d", $db->escapeString($regexMatches['name']), $relparts[0], $relparts[1], Releases::PROCSTAT_TITLEMATCHED, $regexMatches['regcatid'], $regexMatches['regexID'], $db->escapeString($regexMatches['reqID']), $rowbin["ID"])); } else { if ($rowbin['procstat'] == Releases::PROCSTAT_NEW) { $newUnmatchedBinaries[] = $rowbin['ID']; } } } //mark as not matched if (!empty($newUnmatchedBinaries)) { $db->exec(sprintf("update binaries set procstat=%d where ID IN (%s)", Releases::PROCSTAT_TITLENOTMATCHED, implode(',', $newUnmatchedBinaries))); } } // // Move all binaries from releases which have the correct number of files on to the next stage. // echo "Stage 2 : Marking binaries where all parts are available"; $result = $db->queryDirect(sprintf("SELECT relname, date, SUM(reltotalpart) AS reltotalpart, groupID, reqID, fromname, SUM(num) AS num, coalesce(g.minfilestoformrelease, s.minfilestoformrelease) as minfilestoformrelease FROM ( SELECT relname, reltotalpart, groupID, reqID, fromname, max(date) as date, COUNT(ID) AS num FROM binaries WHERE procstat = %s GROUP BY relname, reltotalpart, groupID, reqID, fromname ORDER BY NULL ) x left outer join groups g on g.ID = x.groupID inner join ( select value as minfilestoformrelease from site where setting = 'minfilestoformrelease' ) s GROUP BY relname, groupID, reqID, fromname, minfilestoformrelease ORDER BY NULL", Releases::PROCSTAT_TITLEMATCHED)); while ($row = $db->getAssocArray($result)) { $retcount++; // // Less than the site permitted number of files in a release. Dont discard it, as it may // be part of a set being uploaded. // if ($row["num"] < $row["minfilestoformrelease"]) { //echo "Number of files in release ".$row["relname"]." less than site/group setting (".$row['num']."/".$row["minfilestoformrelease"].")\n"; //$db->exec(sprintf("update binaries set procattempts = procattempts + 1 where relname = %s and procstat = %d and groupID = %d and fromname = %s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]) )); } elseif ($row["num"] >= $row["reltotalpart"]) { $incomplete = false; if ($row['reltotalpart'] == 0 && strtotime($currTime_ori['now']) - strtotime($row['date']) < 14400) { $incomplete = true; } else { // Check that the binary is complete $binlist = $db->query(sprintf("SELECT binaries.ID, totalParts, date, COUNT(DISTINCT parts.messageID) AS num FROM binaries, parts WHERE binaries.ID=parts.binaryID AND binaries.relname = %s AND binaries.procstat = %d AND binaries.groupID = %d AND binaries.fromname = %s GROUP BY binaries.ID ORDER BY NULL", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); foreach ($binlist as $rowbin) { if ($rowbin['num'] < $rowbin['totalParts']) { // Allow to binary to release if posted to usenet longer than four hours ago and we still don't have all the parts if (!(strtotime($currTime_ori['now']) - strtotime($rowbin['date']) > 14400)) { $incomplete = true; break; } } } } if (!$incomplete) { // // Right number of files, but see if the binary is a allfilled/reqid post, in which case it needs its name looked up // if ($row['reqID'] != '' && $page->site->reqidurl != "") { // // Try and get the name using the group // $binGroup = $db->queryOneRow(sprintf("SELECT name FROM groups WHERE ID = %d", $row["groupID"])); $newtitle = $this->getReleaseNameForReqId($page->site->reqidurl, $page->site->newznabID, $binGroup["name"], $row["reqID"]); // // if the feed/group wasnt supported by the scraper, then just use the release name as the title. // if ($newtitle == "no feed") { $newtitle = $row["relname"]; } // // Valid release with right number of files and title now, so move it on // if ($newtitle != "") { $db->exec(sprintf("update binaries set relname = %s, procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($newtitle), Releases::PROCSTAT_READYTORELEASE, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } else { // // Item not found, if the binary was added to the index yages ago, then give up. // $maxaddeddate = $db->queryOneRow(sprintf("SELECT NOW() as now, MAX(dateadded) as dateadded FROM binaries WHERE relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); // // If added to the index over 48 hours ago, give up trying to determine the title // if (strtotime($maxaddeddate['now']) - strtotime($maxaddeddate['dateadded']) > 60 * 60 * 48) { $db->exec(sprintf("update binaries set procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_NOREQIDNAMELOOKUPFOUND, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } } } else { $db->exec(sprintf("update binaries set procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_READYTORELEASE, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } } } if ($retcount % 100 == 0) { echo "."; } } $retcount = 0; echo "\nStage 3 : Creating releases from complete binaries\n"; // // Get out all distinct relname, group from binaries of STAGE2 // $result = $db->queryDirect(sprintf("SELECT relname, groupID, g.name as group_name, fromname, max(categoryID) as categoryID, max(regexID) as regexID, max(reqID) as reqID, MAX(date) as date, count(binaries.ID) as parts, coalesce(sum(binaries.size),0) as size from binaries inner join groups g on g.ID = binaries.groupID where procstat = %d and relname is not null group by relname, g.name, groupID, fromname ORDER BY COUNT(binaries.ID) desc", Releases::PROCSTAT_READYTORELEASE)); while ($row = $db->getAssocArray($result)) { $relguid = md5(uniqid()); // // Get categoryID if one has been allocated to this // if ($row["categoryID"] != "") { $catId = $row["categoryID"]; } else { $catId = $cat->determineCategory($row["group_name"], $row["relname"]); } // // Determine if size matches permitted boundaries and discard here if not. // $gsize = $groupbasedminsizes[$row["groupID"]][0]["minsizetoformrelease"]; if ($gsize == "" || $gsize == 0) { $gsize = $row["size"]; } $ssize = $page->site->minsizetoformrelease; if ($ssize == "" || $ssize == 0) { $ssize = $row["size"]; } $csize = $catbasedsizes[$catId][0]["minsizetoformrelease"]; if ($csize == "" || $csize == 0) { $csize = $row["size"]; } $cpsize = $catbasedsizes[$catId][0]["parentminsizetoformrelease"]; if ($cpsize == "" || $cpsize == 0) { $cpsize = $row["size"]; } $cmaxsize = $catbasedsizes[$catId][0]["maxsizetoformrelease"]; if ($cmaxsize == "" || $cmaxsize == 0) { $cmaxsize = $row["size"]; } $cpmaxsize = $catbasedsizes[$catId][0]["parentmaxsizetoformrelease"]; if ($cpmaxsize == "" || $cpmaxsize == 0) { $cpmaxsize = $row["size"]; } $overallminsize = max($gsize, $ssize, $csize, $cpsize); $overallmaxsize = min($cmaxsize, $cpmaxsize); if ($row["size"] < $overallminsize || $row["size"] > $overallmaxsize) { echo sprintf("Stage 3 : Discarding - %s (Size %s outside permitted range of %s%s)\n", $row["relname"], formatBytes($row["size"]), $overallminsize != $row["size"] ? formatBytes($overallminsize) : "", $overallmaxsize != $row["size"] ? formatBytes($overallmaxsize) : ""); $db->exec(sprintf("DELETE parts, binaries FROM parts JOIN binaries ON binaries.ID = parts.binaryID WHERE relname=%s and groupID=%d and fromname=%s and procstat=%d", $db->escapeString($row["relname"]), $row["groupID"], $db->escapeString($row["fromname"]), Releases::PROCSTAT_READYTORELEASE)); } else { // Clean release name $cleanRelName = $this->cleanReleaseName($row['relname']); $relid = $this->insertRelease($cleanRelName, $row["parts"], $row["groupID"], $relguid, $catId, $row["regexID"], $row["date"], $row["fromname"], $row["reqID"], $page->site); // // Tag every binary for this release with its parent release id // $db->exec(sprintf("update binaries set procstat = %d, releaseID = %d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_RELEASED, $relid, $db->escapeString($row["relname"]), Releases::PROCSTAT_READYTORELEASE, $row["groupID"], $db->escapeString($row["fromname"]))); // // Write the nzb to disk // $nzbfile = $nzb->getNZBPath($relguid, $page->site->nzbpath, true); $nzb->writeNZBforReleaseId($relid, $cleanRelName, $catId, $nzbfile); // // Remove used binaries // $db->exec(sprintf("DELETE parts, binaries FROM parts JOIN binaries ON binaries.ID = parts.binaryID WHERE releaseID = %d ", $relid)); // // If nzb successfully written, then load it and get size completion from it // $nzbInfo = new nzbInfo(); if (!$nzbInfo->loadFromFile($nzbfile)) { echo "Stage 3 : Failed to write nzb file (bad perms?) " . $nzbfile . "\n"; //copy($nzbfile, "./ERRORNZB_".$relguid); $this->delete($relid); } else { // Check if gid already exists $dupes = $db->queryOneRow(sprintf("SELECT EXISTS(SELECT 1 FROM releases WHERE gid = %s) as total", $db->escapeString($nzbInfo->gid))); if ($dupes['total'] > 0) { echo "Stage 3 : Duplicate - " . $cleanRelName . " -" . $nzbInfo->gid . "-\n"; $this->delete($relid); } else { $db->exec(sprintf("update releases set totalpart = %d, size = %s, completion = %d, GID=%s where ID = %d", $nzbInfo->filecount, $nzbInfo->filesize, $nzbInfo->completion, $db->escapeString($nzbInfo->gid), $relid)); echo "Stage 3 : Added release " . $cleanRelName . "\n"; //Increment new release count $retcount++; } } } } // // Delete any releases under the minimum completion percent. // if ($page->site->completionpercent != 0) { echo "Stage 4 : Deleting releases less than " . $page->site->completionpercent . " complete\n"; $result = $db->query(sprintf("select ID from releases where completion > 0 and completion < %d", $page->site->completionpercent)); foreach ($result as $row) { $this->delete($row["ID"]); } } /* *Potentially leave this in to mop up release when the cat sizes change. // // Delete releases whos minsize is less than the site or group minimum // $result = $db->query("select releases.ID from releases left outer join (SELECT g.ID, coalesce(g.minsizetoformrelease, s.minsizetoformrelease) as minsizetoformrelease FROM groups g inner join ( select value as minsizetoformrelease from site where setting = 'minsizetoformrelease' ) s ) x on x.ID = releases.groupID where minsizetoformrelease != 0 and releases.size < minsizetoformrelease"); if (count($result) > 0) { echo "Stage 4 : Deleting ".count($result)." release(s) where size is smaller than minsize for site/group\n"; foreach ($result as $row) $this->delete($row["ID"]); } $result = $db->query("select releases.ID, name, categoryID, size FROM releases JOIN ( select catc.ID, case when catc.minsizetoformrelease = 0 then catp.minsizetoformrelease else catc.minsizetoformrelease end as minsizetoformrelease, case when catc.maxsizetoformrelease = 0 then catp.maxsizetoformrelease else catc.maxsizetoformrelease end as maxsizetoformrelease from category catp join category catc on catc.parentID = catp.ID where (catc.minsizetoformrelease != 0 or catc.maxsizetoformrelease != 0) or (catp.minsizetoformrelease != 0 or catp.maxsizetoformrelease != 0) ) x on x.ID = releases.categoryID where (size < minsizetoformrelease and minsizetoformrelease != 0) or (size > maxsizetoformrelease and maxsizetoformrelease != 0)"); if(count($result) > 0) { echo "Stage 4 : Deleting release(s) not matching category min/max size ...\n"; foreach ($result as $r){ $this->delete($r['ID']); } } */ echo "Stage 5 : Post processing started\n"; $postprocess = new PostProcess(true); $postprocess->processAll(); // // aggregate the releasefiles upto the releases. // echo "Stage 6 : Aggregating Files\n"; $db->exec("update releases INNER JOIN (SELECT releaseID, COUNT(ID) AS num FROM releasefiles GROUP BY releaseID) b ON b.releaseID = releases.ID and releases.rarinnerfilecount = 0 SET rarinnerfilecount = b.num"); // Remove the binaries and parts used to form releases, or that are duplicates. // if ($page->site->partsdeletechunks > 0) { echo "Stage 7 : Chunk deleting unused binaries and parts"; $query = sprintf("SELECT parts.ID as partsID,binaries.ID as binariesID FROM parts\r\n\t\t\t\t\t\tLEFT JOIN binaries ON binaries.ID = parts.binaryID\r\n\t\t\t\t\t\tWHERE binaries.dateadded < %s - INTERVAL %d HOUR LIMIT 0,%d", $db->escapeString($currTime_ori["now"]), ceil($page->site->rawretentiondays * 24), $page->site->partsdeletechunks); $cc = 0; $done = false; while (!$done) { $dd = $cc; $result = $db->query($query); if (count($result) > 0) { $pID = array(); $bID = array(); foreach ($result as $row) { $pID[] = $row['partsID']; $bID[] = $row['binariesID']; } $pID = '(' . implode(',', $pID) . ')'; $bID = '(' . implode(',', $bID) . ')'; $fr = $db->exec("DELETE FROM parts WHERE ID IN {$pID}"); if ($fr > 0) { $cc += $fr; $cc += $db->exec("DELETE FROM binaries WHERE ID IN {$bID}"); } unset($pID); unset($bID); if ($cc == $dd) { $done = true; } echo $cc % 10000 ? '.' : ''; } else { $done = true; } } echo "\nStage 7 : Complete - " . $cc . " rows affected\n"; } else { echo "Stage 7 : Deleting unused binaries and parts\n"; $db->exec(sprintf("DELETE parts, binaries FROM parts JOIN binaries ON binaries.ID = parts.binaryID\r\n\t\t\tWHERE binaries.dateadded < %s - INTERVAL %d HOUR", $db->escapeString($currTime_ori["now"]), ceil($page->site->rawretentiondays * 24))); } // // User/Request housekeeping, should ideally move this to its own section, but it needs to be done automatically. // $users = new Users(); $users->pruneRequestHistory($page->site->userdownloadpurgedays); echo "Done : Added " . $retcount . " releases\n\n"; return $retcount; }
function processReleases() { $db = new DB(); $cat = new Category(); $bin = new Binaries(); $nzb = new Nzb(); $s = new Sites(); $relreg = new ReleaseRegex(); $page = new Page(); $nfo = new Nfo(true); $retcount = 0; echo $s->getLicense(); echo "\n\nStarting release update process (" . date("Y-m-d H:i:s") . ")\n"; if (!file_exists($page->site->nzbpath)) { echo "Bad or missing nzb directory - " . $page->site->nzbpath; return; } $this->checkRegexesUptoDate($page->site->latestregexurl, $page->site->latestregexrevision, true); // // Get all regexes for all groups which are to be applied to new binaries // in order of how they should be applied // $regexrows = $relreg->get(); foreach ($regexrows as $regexrow) { echo "Applying regex " . $regexrow["ID"] . " for group " . ($regexrow["groupname"] == "" ? "all" : $regexrow["groupname"]) . "\n"; $groupmatch = ""; // // Groups ending in * need to be like matched when getting out binaries for groups and children // if (preg_match("/\\*\$/i", $regexrow["groupname"])) { $groupname = substr($regexrow["groupname"], 0, -1); $resgrps = $db->query(sprintf("select ID from groups where name like %s ", $db->escapeString($groupname . "%"))); foreach ($resgrps as $resgrp) { $groupmatch .= " groupID = " . $resgrp["ID"] . " or "; } $groupmatch .= " 1=2 "; } elseif ($regexrow["groupname"] != "") { $resgrp = $db->queryOneRow(sprintf("select ID from groups where name = %s ", $db->escapeString($regexrow["groupname"]))); // // if group not found, its a regex for a group we arent indexing. // if ($resgrp) { $groupmatch = " groupID = " . $resgrp["ID"]; } else { $groupmatch = " 1=2 "; } } else { $groupmatch = " 1=1 "; } // Get current mysql time for date comparison checks in case php is in a different time zone $currTime = $db->queryOneRow("SELECT NOW() as now"); // Get out all binaries of STAGE0 for current group $arrNoPartBinaries = array(); $resbin = $db->queryDirect(sprintf("SELECT binaries.ID, binaries.name, binaries.date, binaries.totalParts from binaries where (%s) and procstat = %d order by binaries.date asc", $groupmatch, Releases::PROCSTAT_NEW)); while ($rowbin = mysql_fetch_assoc($resbin)) { if (preg_match($regexrow["regex"], $rowbin["name"], $matches)) { $matches = array_map("trim", $matches); if (isset($matches['reqid']) && ctype_digit($matches['reqid']) && (!isset($matches['name']) || empty($matches['name']))) { $matches['name'] = $matches['reqid']; } // Check that the regex provided the correct parameters if (!isset($matches['name']) || empty($matches['name'])) { echo "regex applied which didnt return right number of capture groups - " . $regexrow["regex"] . "\n"; print_r($matches); continue; } // If theres no number of files data in the subject, put it into a release if it was posted to usenet longer than five hours ago. if (!isset($matches['parts']) && strtotime($currTime['now']) - strtotime($rowbin['date']) > 18000 || isset($arrNoPartBinaries[$matches['name']])) { // // Take a copy of the name of this no-part release found. This can be used // next time round the loop to find parts of this set, but which have not yet reached 3 hours. // $arrNoPartBinaries[$matches['name']] = "1"; $matches['parts'] = "01/01"; } if (isset($matches['name']) && isset($matches['parts'])) { if (strpos($matches['parts'], '/') === false) { $matches['parts'] = str_replace(array('-', '~', ' of '), '/', $matches['parts']); } $regcatid = "null "; if ($regexrow["categoryID"] != "") { $regcatid = $regexrow["categoryID"]; } $reqid = " null "; if (isset($matches['reqid'])) { $reqid = $matches['reqid']; } //check if post is repost if (preg_match('/(repost\\d?|re\\-?up)/i', $rowbin['name'], $repost) && !preg_match('/repost|re\\-?up/i', $matches['name'])) { $matches['name'] .= ' ' . $repost[1]; } $relparts = explode("/", $matches['parts']); $db->query(sprintf("update binaries set relname = replace(%s, '_', ' '), relpart = %d, reltotalpart = %d, procstat=%d, categoryID=%s, regexID=%d, reqID=%s where ID = %d", $db->escapeString($matches['name']), $relparts[0], $relparts[1], Releases::PROCSTAT_TITLEMATCHED, $regcatid, $regexrow["ID"], $reqid, $rowbin["ID"])); } } } } // // Move all binaries from releases which have the correct number of files on to the next stage. // echo "Stage 2\n"; $result = $db->queryDirect(sprintf("SELECT relname, SUM(reltotalpart) AS reltotalpart, groupID, reqID, fromname, SUM(num) AS num, coalesce(g.minfilestoformrelease, s.minfilestoformrelease) as minfilestoformrelease FROM ( SELECT relname, reltotalpart, groupID, reqID, fromname, COUNT(ID) AS num FROM binaries WHERE procstat = %s GROUP BY relname, reltotalpart, groupID, reqID, fromname ) x left outer join groups g on g.ID = x.groupID inner join ( select value as minfilestoformrelease from site where setting = 'minfilestoformrelease' ) s GROUP BY relname, groupID, reqID, fromname", Releases::PROCSTAT_TITLEMATCHED)); while ($row = mysql_fetch_assoc($result)) { $retcount++; // // Less than the site permitted number of files in a release. Dont discard it, as it may // be part of a set being uploaded. // if ($row["num"] < $row["minfilestoformrelease"]) { echo "Number of files in release " . $row["relname"] . " less than site/group setting (" . $row['num'] . "/" . $row["minfilestoformrelease"] . ")\n"; $db->query(sprintf("update binaries set procattempts = procattempts + 1 where relname = %s and procstat = %d and groupID = %d and fromname = %s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } elseif ($row["num"] >= $row["reltotalpart"]) { // Check that the binary is complete $binlist = $db->query(sprintf("SELECT ID, totalParts, date from binaries where relname = %s and procstat = %d and groupID = %d and fromname = %s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); $incomplete = false; foreach ($binlist as $rowbin) { $binParts = $db->queryOneRow(sprintf("SELECT COUNT(ID) AS num FROM parts WHERE binaryID = %d", $rowbin['ID'])); if ($binParts['num'] < $rowbin['totalParts']) { echo "binary " . $rowbin['ID'] . " from " . $row['relname'] . " has missing parts - " . $binParts['num'] . "/" . $rowbin['totalParts'] . " (" . number_format($binParts['num'] / $rowbin['totalParts'] * 100, 1) . "% complete)\n"; // Allow to binary to release if posted to usenet longer than four hours ago and we still don't have all the parts if (strtotime($currTime['now']) - strtotime($rowbin['date']) > 14400) { echo "allowing incomplete binary " . $rowbin['ID'] . "\n"; } else { $incomplete = true; } } } if ($incomplete) { echo "Incorrect number of parts " . $row["relname"] . "-" . $row["num"] . "-" . $row["reltotalpart"] . "\n"; $db->query(sprintf("update binaries set procattempts = procattempts + 1 where relname = %s and procstat = %d and groupID = %d and fromname = %s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } elseif ($row['reqID'] != '' && $page->site->reqidurl != "") { // // Try and get the name using the group // $binGroup = $db->queryOneRow(sprintf("SELECT name FROM groups WHERE ID = %d", $row["groupID"])); echo "Looking up " . $row['reqID'] . " in " . $binGroup['name'] . "... "; $newtitle = $this->getReleaseNameForReqId($page->site->reqidurl, $binGroup["name"], $row["reqID"], true); // // if the feed/group wasnt supported by the scraper, then just use the release name as the title. // if ($newtitle == "no feed") { $newtitle = $row["relname"]; echo "Group not supported\n"; } // // Valid release with right number of files and title now, so move it on // if ($newtitle != "") { $db->query(sprintf("update binaries set relname = %s, procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($newtitle), Releases::PROCSTAT_READYTORELEASE, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } else { // // Item not found, if the binary was added to the index yages ago, then give up. // $maxaddeddate = $db->queryOneRow(sprintf("SELECT NOW() as now, MAX(dateadded) as dateadded FROM binaries WHERE relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); // // If added to the index over 48 hours ago, give up trying to determine the title // if (strtotime($maxaddeddate['now']) - strtotime($maxaddeddate['dateadded']) > 60 * 60 * 48) { $db->query(sprintf("update binaries set procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_NOREQIDNAMELOOKUPFOUND, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); echo "Not found in 48 hours\n"; } } } else { $db->query(sprintf("update binaries set procstat=%d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_READYTORELEASE, $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } } else { echo "Incorrect number of files for " . $row["relname"] . " (" . $row["num"] . "/" . $row["reltotalpart"] . ")\n"; $db->query(sprintf("update binaries set procattempts = procattempts + 1 where relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($row["relname"]), Releases::PROCSTAT_TITLEMATCHED, $row["groupID"], $db->escapeString($row["fromname"]))); } if ($retcount % 10 == 0) { echo "-processed " . $retcount . " binaries stage two\n"; } } $retcount = $nfocount = 0; echo "Stage 3\n"; // // Get out all distinct relname, group from binaries of STAGE2 // $result = $db->queryDirect(sprintf("SELECT relname, groupID, g.name as group_name, fromname, count(binaries.ID) as parts from binaries inner join groups g on g.ID = binaries.groupID where procstat = %d and relname is not null group by relname, g.name, groupID, fromname ORDER BY COUNT(binaries.ID) desc", Releases::PROCSTAT_READYTORELEASE)); while ($row = mysql_fetch_assoc($result)) { $retcount++; // // Get the last post date and the poster name from the binary // $bindata = $db->queryOneRow(sprintf("select fromname, MAX(date) as date from binaries where relname = %s and procstat = %d and groupID = %d and fromname = %s group by fromname", $db->escapeString($row["relname"]), Releases::PROCSTAT_READYTORELEASE, $row["groupID"], $db->escapeString($row["fromname"]))); // // Get all releases with the same name with a usenet posted date in a +1-1 date range. // $relDupes = $db->query(sprintf("select ID from releases where searchname = %s and (%s - INTERVAL 1 DAY < postdate AND %s + INTERVAL 1 DAY > postdate)", $db->escapeString($row["relname"]), $db->escapeString($bindata["date"]), $db->escapeString($bindata["date"]))); if (count($relDupes) > 0) { $db->query(sprintf("update binaries set procstat = %d where relname = %s and procstat = %d and groupID = %d and fromname=%s ", Releases::PROCSTAT_DUPLICATE, $db->escapeString($row["relname"]), Releases::PROCSTAT_READYTORELEASE, $row["groupID"], $db->escapeString($row["fromname"]))); continue; } // // Get total size of this release // Done in a big OR statement, not an IN as the mysql binaryID index on parts table // was not being used. // $totalSize = "0"; $regexAppliedCategoryID = ""; $regexIDused = ""; $reqIDused = ""; $relTotalParts = 0; $relCompletion = 0; $binariesForSize = $db->query(sprintf("select ID, categoryID, regexID, reqID, totalParts from binaries use index (ix_binary_relname) where relname = %s and procstat = %d and groupID = %d and fromname=%s", $db->escapeString($row["relname"]), Releases::PROCSTAT_READYTORELEASE, $row["groupID"], $db->escapeString($row["fromname"]))); if (count($binariesForSize) > 0) { $sizeSql = "select sum(size) as totalSize, count(ID) as relParts from parts where ("; foreach ($binariesForSize as $binSizeId) { $sizeSql .= " binaryID = " . $binSizeId["ID"] . " or "; // // Get categoryID if one has been allocated to this // if ($binSizeId["categoryID"] != "" && $regexAppliedCategoryID == "") { $regexAppliedCategoryID = $binSizeId["categoryID"]; } // // Get RegexID if one has been allocated to this // if ($binSizeId["regexID"] != "" && $regexIDused == "") { $regexIDused = $binSizeId["regexID"]; } // // Get requestID if one has been allocated to this // if ($binSizeId["reqID"] != "" && $reqIDused == "") { $reqIDused = $binSizeId["reqID"]; } // // Get number of expected parts // $relTotalParts += $binSizeId["totalParts"]; } $sizeSql .= " 1=2) "; $temp = $db->queryOneRow($sizeSql); $totalSize = $temp["totalSize"] + 0 . ""; $relCompletion = number_format($temp["relParts"] / $relTotalParts * 100, 1, '.', ''); } // // Insert the release // $relguid = md5(uniqid()); if ($regexAppliedCategoryID == "") { $catId = $cat->determineCategory($row["group_name"], $row["relname"]); } else { $catId = $regexAppliedCategoryID; } if ($regexIDused == "") { $regexID = " null "; } else { $regexID = $regexIDused; } if ($reqIDused == "") { $reqID = " null "; } else { $reqID = $reqIDused; } //Clean release name $cleanArr = array('#', '@', '$', '%', '^', '§', '¨', '©', 'Ö'); $cleanRelName = str_replace($cleanArr, '', $row['relname']); $relid = $db->queryInsert(sprintf("insert into releases (name, searchname, totalpart, groupID, adddate, guid, categoryID, regexID, rageID, postdate, fromname, size, reqID, passwordstatus, completion) values (%s, %s, %d, %d, now(), %s, %d, %d, -1, %s, %s, %s, %s, %d, %f)", $db->escapeString($cleanRelName), $db->escapeString($cleanRelName), $row["parts"], $row["groupID"], $db->escapeString($relguid), $catId, $regexID, $db->escapeString($bindata["date"]), $db->escapeString($bindata["fromname"]), $totalSize, $reqID, $page->site->checkpasswordedrar == "1" ? -1 : 0, $relCompletion > 100 ? 100 : $relCompletion)); echo "Added release " . $cleanRelName . "\n"; // // Tag every binary for this release with its parent release id // remove the release name from the binary as its no longer required // $db->query(sprintf("update binaries set procstat = %d, releaseID = %d where relname = %s and procstat = %d and groupID = %d and fromname=%s", Releases::PROCSTAT_RELEASED, $relid, $db->escapeString($row["relname"]), Releases::PROCSTAT_READYTORELEASE, $row["groupID"], $db->escapeString($row["fromname"]))); // // Find an .nfo in the release // $relnfo = $nfo->determineReleaseNfo($relid); if ($relnfo !== false) { $nfo->addReleaseNfo($relid, $relnfo['ID']); $nfocount++; } // // Write the nzb to disk // $nzb->writeNZBforReleaseId($relid, $relguid, $cleanRelName, $catId, $nzb->getNZBPath($relguid, $page->site->nzbpath, true)); if ($retcount % 5 == 0) { echo "-processed " . $retcount . " releases stage three\n"; } } echo "Found " . $nfocount . " nfos in " . $retcount . " releases\n"; // // Process nfo files // if ($page->site->lookupnfo != "1") { echo "Site config (site.lookupnfo) prevented retrieving nfos\n"; } else { $nfo->processNfoFiles($page->site->lookupimdb, $page->site->lookuptvrage == "1"); } // // Lookup imdb if enabled // if ($page->site->lookupimdb == 1) { $movie = new Movie(true); $movie->processMovieReleases(); } // // Lookup music if enabled // if ($page->site->lookupmusic == 1) { $music = new Music(true); $music->processMusicReleases(); } // // Lookup games if enabled // if ($page->site->lookupgames == 1) { $console = new Console(true); $console->processConsoleReleases(); } // // Check for passworded releases // if ($page->site->checkpasswordedrar != "1") { echo "Site config (site.checkpasswordedrar) prevented checking releases are passworded\n"; } else { $this->processPasswordedReleases(true); } // // Process all TV related releases which will assign their series/episode/rage data // $tvrage = new TVRage(true); $tvrage->processTvReleases($page->site->lookuptvrage == "1"); // // Get the current datetime again, as using now() in the housekeeping queries prevents the index being used. // $currTime = $db->queryOneRow("SELECT NOW() as now"); // // Tidy away any binaries which have been attempted to be grouped into // a release more than x times // echo "Tidying away binaries which cant be grouped after " . $page->site->attemptgroupbindays . " days\n"; $db->query(sprintf("update binaries set procstat = %d where procstat = %d and dateadded < %s - interval %d day ", Releases::PROCSTAT_WRONGPARTS, Releases::PROCSTAT_NEW, $db->escapeString($currTime["now"]), $page->site->attemptgroupbindays)); // // Delete any parts and binaries which are older than the site's retention days // echo "Deleting parts which are older than " . $page->site->rawretentiondays . " days\n"; $db->query(sprintf("delete from parts where dateadded < %s - interval %d day", $db->escapeString($currTime["now"]), $page->site->rawretentiondays)); echo "Deleting binaries which are older than " . $page->site->rawretentiondays . " days\n"; $db->query(sprintf("delete from binaries where dateadded < %s - interval %d day", $db->escapeString($currTime["now"]), $page->site->rawretentiondays)); // // Delete any releases which are older than site's release retention days // if ($page->site->releaseretentiondays != 0) { echo "Determining any releases past retention to be deleted.\n\n"; $result = $db->query(sprintf("select ID from releases where postdate < %s - interval %d day", $db->escapeString($currTime["now"]), $page->site->releaseretentiondays)); foreach ($result as $row) { $this->delete($row["ID"]); } } echo "Processed " . $retcount . " releases\n\n"; return $retcount; }
define('FS_ROOT', realpath(dirname(__FILE__))); require_once FS_ROOT . "/../../www/config.php"; require_once FS_ROOT . "/../../www/lib/framework/db.php"; require_once FS_ROOT . "/../../www/lib/category.php"; $db = new DB(); $category = new Category(); $changedcount = 0; $rescount = 0; // // [1] change for all where the regex isnt aimed at a specific category (affects most releases) // //$res = $db->queryDirect("select r.id as id, r.searchname as searchname, g.name as groupname, r.categoryID as rcat from releases r, releaseregex rr, groups g where r.regexid=rr.id and g.ID = r.groupID and rr.categoryid is null"); // // [2] update for all in a category // //$res = $db->queryDirect("select r.id as id, r.searchname as searchname, g.name as groupname, r.categoryID as rcat from releases r inner join groups g on g.ID = r.groupID where r.categoryID in (6010, 6020, 6030, 6040)"); // // [3] reset all flac with hashed names // //$res = $db->queryDirect("select releases.ID as id, searchname as searchname, groups.name as groupname, releases.categoryID as rcat from releases join groups on groups.ID = releases.groupID where length(searchname) = 40 and groups.name like 'alt.binaries.sounds.flac'"); while ($rel = $db->getAssocArray($res)) { $rescount++; $categoryID = $category->determineCategory($rel['groupname'], $rel['searchname']); if ($categoryID != $rel['rcat'] && $categoryID != '7900') { $changedcount++; echo "Changing category for " . $rel['searchname'] . " New (" . $categoryID . ") Old (" . $rel['rcat'] . ")\n"; $db->exec(sprintf("update releases SET categoryID = %d WHERE ID = %d", $categoryID, $rel['id'])); } } echo $rescount . " releases \n"; echo $changedcount . " releases changed\n";