up pages
This commit is contained in:
Tykayn
tykayn
parent
dffb21b56e
commit
1ed74c2e2f
9 changed files with 1610 additions and 305 deletions
464
test_decrepitude.php
Normal file
464
test_decrepitude.php
Normal file
|
|
@ -0,0 +1,464 @@
|
|||
<?php
|
||||
|
||||
// Set memory limit to match the error condition (128MB)
|
||||
ini_set('memory_limit', '128M');
|
||||
|
||||
// Path to the large JSON file
|
||||
$outdatedPagesFile = __DIR__ . '/wiki_compare/outdated_pages.json';
|
||||
$histogramFile = __DIR__ . '/wiki_compare/staleness_histogram.png';
|
||||
|
||||
echo "Testing memory usage for decrepitudeScores() method\n";
|
||||
echo "File size: " . round(filesize($outdatedPagesFile) / (1024 * 1024), 2) . " MB\n";
|
||||
echo "Memory limit: " . ini_get('memory_limit') . "\n\n";
|
||||
|
||||
// Test the original approach (loading entire file)
|
||||
function testOriginalApproach($filePath) {
|
||||
echo "Testing original approach (loading entire file)...\n";
|
||||
$memBefore = memory_get_usage();
|
||||
echo "Memory usage before: " . round($memBefore / (1024 * 1024), 2) . " MB\n";
|
||||
|
||||
try {
|
||||
$startTime = microtime(true);
|
||||
|
||||
$regularPages = [];
|
||||
$specificPages = [];
|
||||
$lastUpdated = null;
|
||||
|
||||
if (file_exists($filePath)) {
|
||||
$outdatedPagesData = json_decode(file_get_contents($filePath), true);
|
||||
|
||||
if (isset($outdatedPagesData['regular_pages']) && is_array($outdatedPagesData['regular_pages'])) {
|
||||
$regularPages = $outdatedPagesData['regular_pages'];
|
||||
}
|
||||
|
||||
if (isset($outdatedPagesData['specific_pages']) && is_array($outdatedPagesData['specific_pages'])) {
|
||||
$specificPages = $outdatedPagesData['specific_pages'];
|
||||
}
|
||||
|
||||
if (isset($outdatedPagesData['last_updated'])) {
|
||||
$lastUpdated = $outdatedPagesData['last_updated'];
|
||||
}
|
||||
}
|
||||
|
||||
$endTime = microtime(true);
|
||||
|
||||
echo "Successfully loaded data:\n";
|
||||
echo "- Regular pages: " . count($regularPages) . "\n";
|
||||
echo "- Specific pages: " . count($specificPages) . "\n";
|
||||
echo "- Last updated: " . ($lastUpdated ?? 'null') . "\n";
|
||||
|
||||
$memAfter = memory_get_usage();
|
||||
echo "Memory usage after: " . round($memAfter / (1024 * 1024), 2) . " MB\n";
|
||||
echo "Memory used: " . round(($memAfter - $memBefore) / (1024 * 1024), 2) . " MB\n";
|
||||
echo "Time taken: " . round($endTime - $startTime, 2) . " seconds\n";
|
||||
} catch (Exception $e) {
|
||||
echo "Error: " . $e->getMessage() . "\n";
|
||||
}
|
||||
|
||||
echo "\n";
|
||||
}
|
||||
|
||||
// Test the new approach (streaming)
|
||||
function testNewApproach($filePath) {
|
||||
echo "Testing new approach (streaming)...\n";
|
||||
$memBefore = memory_get_usage();
|
||||
echo "Memory usage before: " . round($memBefore / (1024 * 1024), 2) . " MB\n";
|
||||
|
||||
try {
|
||||
$startTime = microtime(true);
|
||||
|
||||
$regularPages = extractJsonArrayByKey($filePath, 'regular_pages', 100);
|
||||
$specificPages = extractJsonArrayByKey($filePath, 'specific_pages', 100);
|
||||
$lastUpdated = extractJsonScalarByKey($filePath, 'last_updated');
|
||||
|
||||
$endTime = microtime(true);
|
||||
|
||||
echo "Successfully loaded data:\n";
|
||||
echo "- Regular pages: " . count($regularPages) . "\n";
|
||||
echo "- Specific pages: " . count($specificPages) . "\n";
|
||||
echo "- Last updated: " . ($lastUpdated ?? 'null') . "\n";
|
||||
|
||||
$memAfter = memory_get_usage();
|
||||
echo "Memory usage after: " . round($memAfter / (1024 * 1024), 2) . " MB\n";
|
||||
echo "Memory used: " . round(($memAfter - $memBefore) / (1024 * 1024), 2) . " MB\n";
|
||||
echo "Time taken: " . round($endTime - $startTime, 2) . " seconds\n";
|
||||
} catch (Exception $e) {
|
||||
echo "Error: " . $e->getMessage() . "\n";
|
||||
}
|
||||
|
||||
echo "\n";
|
||||
}
|
||||
|
||||
// Implementation of extractJsonArrayByKey
|
||||
function extractJsonArrayByKey(string $filePath, string $key, int $maxItems = 100): array
|
||||
{
|
||||
$result = [];
|
||||
|
||||
// First, check if the file exists and is readable
|
||||
if (!is_readable($filePath)) {
|
||||
echo "File is not readable: $filePath\n";
|
||||
return $result;
|
||||
}
|
||||
|
||||
// Get the file size
|
||||
$fileSize = filesize($filePath);
|
||||
if ($fileSize === false || $fileSize === 0) {
|
||||
echo "File is empty or size could not be determined: $filePath\n";
|
||||
return $result;
|
||||
}
|
||||
|
||||
try {
|
||||
// For very large files, we'll use a more efficient approach
|
||||
// We'll search for the specified key directly
|
||||
$handle = fopen($filePath, 'r');
|
||||
if (!$handle) {
|
||||
echo "Could not open file: $filePath\n";
|
||||
return $result;
|
||||
}
|
||||
|
||||
// Variables to track parsing state
|
||||
$bracketCount = 0;
|
||||
$buffer = '';
|
||||
$itemCount = 0;
|
||||
$inArray = false;
|
||||
$arrayStarted = false;
|
||||
|
||||
// Skip ahead to find the specified key more quickly
|
||||
$found = false;
|
||||
$searchKey = '"' . $key . '"';
|
||||
$lineCount = 0;
|
||||
|
||||
while (!$found && ($line = fgets($handle)) !== false) {
|
||||
$lineCount++;
|
||||
if ($lineCount % 1000 === 0) {
|
||||
echo "Processed $lineCount lines searching for $key...\r";
|
||||
}
|
||||
|
||||
if (strpos($line, $searchKey) !== false) {
|
||||
$found = true;
|
||||
echo "\nFound $key key at line $lineCount\n";
|
||||
|
||||
// Extract everything after the key
|
||||
$keyPos = strpos($line, $searchKey);
|
||||
$afterKey = substr($line, $keyPos + strlen($searchKey));
|
||||
|
||||
// Find the colon and then the opening bracket
|
||||
if (strpos($afterKey, ':') !== false && strpos($afterKey, '[') !== false) {
|
||||
$inArray = true;
|
||||
$arrayStarted = true;
|
||||
$bracketPos = strpos($afterKey, '[');
|
||||
$buffer = '['; // Start the buffer with an opening bracket
|
||||
$bracketCount = 1;
|
||||
|
||||
// Add everything after the opening bracket to the buffer
|
||||
$buffer .= substr($afterKey, $bracketPos + 1);
|
||||
echo "Opening bracket found on the same line\n";
|
||||
} else if (strpos($afterKey, ':') !== false) {
|
||||
// The opening bracket might be on the next line
|
||||
$inArray = true;
|
||||
echo "Colon found, but opening bracket might be on the next line\n";
|
||||
}
|
||||
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// If we didn't find the key, return empty array
|
||||
if (!$found) {
|
||||
echo "$key key not found in the file\n";
|
||||
fclose($handle);
|
||||
return $result;
|
||||
}
|
||||
|
||||
// If we found the key but not the opening bracket yet, look for it
|
||||
if ($inArray && !$arrayStarted) {
|
||||
echo "Looking for opening bracket...\n";
|
||||
while (($line = fgets($handle)) !== false) {
|
||||
if (strpos($line, '[') !== false) {
|
||||
$bracketPos = strpos($line, '[');
|
||||
$buffer = '['; // Start the buffer with an opening bracket
|
||||
$bracketCount = 1;
|
||||
$arrayStarted = true;
|
||||
|
||||
// Add everything after the opening bracket to the buffer
|
||||
$buffer .= substr($line, $bracketPos + 1);
|
||||
echo "Opening bracket found on the next line\n";
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// If we still haven't found the opening bracket, something is wrong
|
||||
if (!$arrayStarted) {
|
||||
echo "Could not find opening bracket for array '$key' in file: $filePath\n";
|
||||
fclose($handle);
|
||||
return $result;
|
||||
}
|
||||
|
||||
echo "Processing $key array...\n";
|
||||
|
||||
// Now process the array
|
||||
$collectingItems = true;
|
||||
while ($collectingItems && ($line = fgets($handle)) !== false) {
|
||||
// Count opening and closing brackets to track array nesting
|
||||
$openBrackets = substr_count($line, '[') + substr_count($line, '{');
|
||||
$closeBrackets = substr_count($line, ']') + substr_count($line, '}');
|
||||
$bracketCount += $openBrackets - $closeBrackets;
|
||||
|
||||
// Add the line to our buffer
|
||||
$buffer .= $line;
|
||||
|
||||
// If we've reached the end of the array (bracketCount = 0)
|
||||
if ($bracketCount === 0) {
|
||||
$collectingItems = false;
|
||||
echo "Reached end of $key array\n";
|
||||
|
||||
// Try to parse the buffer as JSON
|
||||
try {
|
||||
$parsedData = json_decode($buffer, true);
|
||||
if (json_last_error() !== JSON_ERROR_NONE) {
|
||||
echo "JSON parse error: " . json_last_error_msg() . " for key '$key'\n";
|
||||
// Debug: output a small part of the buffer
|
||||
echo "Buffer preview (first 100 chars): " . substr($buffer, 0, 100) . "...\n";
|
||||
echo "Buffer preview (last 100 chars): ..." . substr($buffer, -100) . "\n";
|
||||
} else if (is_array($parsedData)) {
|
||||
// Limit the number of items to prevent memory exhaustion
|
||||
$result = array_slice($parsedData, 0, $maxItems);
|
||||
echo "Parsed " . count($result) . " items from the $key array\n";
|
||||
}
|
||||
} catch (Exception $e) {
|
||||
echo "Exception parsing JSON for key '$key': " . $e->getMessage() . "\n";
|
||||
}
|
||||
|
||||
break;
|
||||
}
|
||||
|
||||
// Check if we've found a complete item (when we see a closing brace followed by a comma)
|
||||
// This is used to count items and limit the number of items processed
|
||||
if (preg_match('/\}\s*,\s*$/m', $line)) {
|
||||
$itemCount++;
|
||||
if ($itemCount % 10 === 0) {
|
||||
echo "Found $itemCount items in $key array...\r";
|
||||
}
|
||||
|
||||
// If we've reached the maximum number of items, stop processing
|
||||
if ($itemCount >= $maxItems) {
|
||||
$collectingItems = false;
|
||||
echo "\nReached maximum number of items ($maxItems) for $key\n";
|
||||
|
||||
// Create a valid JSON array with the items we've collected so far
|
||||
// We need to ensure the buffer ends with a complete JSON object and a closing bracket
|
||||
|
||||
// First, find the last complete object (ending with })
|
||||
$lastObjectEnd = strrpos($buffer, '}');
|
||||
if ($lastObjectEnd !== false) {
|
||||
// Truncate the buffer at the end of the last complete object
|
||||
$buffer = substr($buffer, 0, $lastObjectEnd + 1);
|
||||
// Add the closing bracket for the array
|
||||
$buffer .= ']';
|
||||
|
||||
echo "Truncated buffer and added closing bracket\n";
|
||||
|
||||
// Try to parse the buffer as JSON
|
||||
try {
|
||||
$parsedData = json_decode($buffer, true);
|
||||
if (json_last_error() !== JSON_ERROR_NONE) {
|
||||
echo "JSON parse error after max items: " . json_last_error_msg() . " for key '$key'\n";
|
||||
// Debug: output a small part of the buffer
|
||||
echo "Buffer preview (first 100 chars): " . substr($buffer, 0, 100) . "...\n";
|
||||
echo "Buffer preview (last 100 chars): ..." . substr($buffer, -100) . "\n";
|
||||
|
||||
// Try a different approach - manually construct a valid JSON array
|
||||
echo "Trying alternative approach to construct valid JSON...\n";
|
||||
|
||||
// Split the buffer by objects (each starting with { and ending with })
|
||||
preg_match_all('/\{[^{}]*(?:\{[^{}]*\}[^{}]*)*\}/s', $buffer, $matches);
|
||||
|
||||
if (!empty($matches[0])) {
|
||||
// Take the first $maxItems objects
|
||||
$objects = array_slice($matches[0], 0, $maxItems);
|
||||
|
||||
// Construct a valid JSON array
|
||||
$validJson = '[' . implode(',', $objects) . ']';
|
||||
|
||||
// Try to parse the valid JSON
|
||||
$parsedData = json_decode($validJson, true);
|
||||
if (json_last_error() === JSON_ERROR_NONE && is_array($parsedData)) {
|
||||
$result = $parsedData;
|
||||
echo "Successfully parsed " . count($result) . " items using alternative approach\n";
|
||||
} else {
|
||||
echo "Alternative approach also failed: " . json_last_error_msg() . "\n";
|
||||
}
|
||||
}
|
||||
} else if (is_array($parsedData)) {
|
||||
$result = $parsedData;
|
||||
echo "Parsed " . count($result) . " items from the $key array\n";
|
||||
}
|
||||
} catch (Exception $e) {
|
||||
echo "Exception parsing JSON after max items for key '$key': " . $e->getMessage() . "\n";
|
||||
}
|
||||
} else {
|
||||
echo "Could not find the end of the last complete object\n";
|
||||
}
|
||||
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Close the file
|
||||
fclose($handle);
|
||||
|
||||
} catch (Exception $e) {
|
||||
echo "Exception in extractJsonArrayByKey for key '$key': " . $e->getMessage() . "\n";
|
||||
}
|
||||
|
||||
return $result;
|
||||
}
|
||||
|
||||
// Implementation of extractJsonScalarByKey
|
||||
function extractJsonScalarByKey(string $filePath, string $key): mixed
|
||||
{
|
||||
// First, check if the file exists and is readable
|
||||
if (!is_readable($filePath)) {
|
||||
echo "File is not readable: $filePath\n";
|
||||
return null;
|
||||
}
|
||||
|
||||
try {
|
||||
// For very large files, we'll use a more efficient approach
|
||||
// We'll search for the specified key directly
|
||||
$handle = fopen($filePath, 'r');
|
||||
if (!$handle) {
|
||||
echo "Could not open file: $filePath\n";
|
||||
return null;
|
||||
}
|
||||
|
||||
// Skip ahead to find the specified key more quickly
|
||||
$found = false;
|
||||
$searchKey = '"' . $key . '"';
|
||||
$value = null;
|
||||
$lineCount = 0;
|
||||
|
||||
while (!$found && ($line = fgets($handle)) !== false) {
|
||||
$lineCount++;
|
||||
if ($lineCount % 1000 === 0) {
|
||||
echo "Processed $lineCount lines searching for $key...\r";
|
||||
}
|
||||
|
||||
if (strpos($line, $searchKey) !== false) {
|
||||
$found = true;
|
||||
echo "\nFound $key key at line $lineCount\n";
|
||||
|
||||
// Extract everything after the key
|
||||
$keyPos = strpos($line, $searchKey);
|
||||
$afterKey = substr($line, $keyPos + strlen($searchKey));
|
||||
|
||||
// Check if the value is on this line
|
||||
if (strpos($afterKey, ':') !== false) {
|
||||
$colonPos = strpos($afterKey, ':');
|
||||
$afterColon = trim(substr($afterKey, $colonPos + 1));
|
||||
|
||||
// Extract the value based on its type
|
||||
if (preg_match('/^"([^"]*)"/', $afterColon, $matches)) {
|
||||
// String value
|
||||
$value = $matches[1];
|
||||
echo "Extracted string value: $value\n";
|
||||
} elseif (preg_match('/^(\d+)/', $afterColon, $matches)) {
|
||||
// Numeric value
|
||||
$value = intval($matches[1]);
|
||||
echo "Extracted numeric value: $value\n";
|
||||
} elseif (preg_match('/^(true|false)/', $afterColon, $matches)) {
|
||||
// Boolean value
|
||||
$value = ($matches[1] === 'true');
|
||||
echo "Extracted boolean value: " . ($value ? 'true' : 'false') . "\n";
|
||||
} elseif (strpos($afterColon, 'null') === 0) {
|
||||
// Null value
|
||||
$value = null;
|
||||
echo "Extracted null value\n";
|
||||
} else {
|
||||
// The value might be on the next line or more complex
|
||||
// For simplicity, we'll just use the regex approach as a fallback
|
||||
echo "Using fallback method to extract value\n";
|
||||
if (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*"([^"]*)"/', $line, $matches)) {
|
||||
// String value
|
||||
$value = $matches[1];
|
||||
echo "Extracted string value: $value\n";
|
||||
} elseif (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*(\d+)/', $line, $matches)) {
|
||||
// Numeric value
|
||||
$value = intval($matches[1]);
|
||||
echo "Extracted numeric value: $value\n";
|
||||
} elseif (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*(true|false)/', $line, $matches)) {
|
||||
// Boolean value
|
||||
$value = ($matches[1] === 'true');
|
||||
echo "Extracted boolean value: " . ($value ? 'true' : 'false') . "\n";
|
||||
} elseif (strpos($line, 'null') !== false) {
|
||||
// Null value
|
||||
$value = null;
|
||||
echo "Extracted null value\n";
|
||||
} else {
|
||||
echo "Could not extract value for key '$key' from line: " . trim($line) . "\n";
|
||||
}
|
||||
}
|
||||
} else {
|
||||
// The value might be on the next line
|
||||
echo "Value for key '$key' might be on the next line, using fallback method\n";
|
||||
|
||||
// Read the next line
|
||||
$nextLine = fgets($handle);
|
||||
if ($nextLine !== false) {
|
||||
$combinedLine = $line . $nextLine;
|
||||
|
||||
// Try to extract the value using regex
|
||||
if (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*"([^"]*)"/', $combinedLine, $matches)) {
|
||||
// String value
|
||||
$value = $matches[1];
|
||||
echo "Extracted string value: $value\n";
|
||||
} elseif (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*(\d+)/', $combinedLine, $matches)) {
|
||||
// Numeric value
|
||||
$value = intval($matches[1]);
|
||||
echo "Extracted numeric value: $value\n";
|
||||
} elseif (preg_match('/"' . preg_quote($key, '/') . '"\s*:\s*(true|false)/', $combinedLine, $matches)) {
|
||||
// Boolean value
|
||||
$value = ($matches[1] === 'true');
|
||||
echo "Extracted boolean value: " . ($value ? 'true' : 'false') . "\n";
|
||||
} elseif (strpos($combinedLine, 'null') !== false) {
|
||||
// Null value
|
||||
$value = null;
|
||||
echo "Extracted null value\n";
|
||||
} else {
|
||||
echo "Could not extract value for key '$key' from combined lines\n";
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// Close the file
|
||||
fclose($handle);
|
||||
|
||||
if (!$found) {
|
||||
echo "Key '$key' not found in file: $filePath\n";
|
||||
} else if ($value === null) {
|
||||
echo "Value for key '$key' is null or could not be extracted\n";
|
||||
}
|
||||
|
||||
return $value;
|
||||
|
||||
} catch (Exception $e) {
|
||||
echo "Exception in extractJsonScalarByKey for key '$key': " . $e->getMessage() . "\n";
|
||||
return null;
|
||||
}
|
||||
}
|
||||
|
||||
// Skip the original approach since we know it fails with memory exhaustion
|
||||
echo "=== ORIGINAL APPROACH ===\n";
|
||||
echo "Skipping original approach - known to fail with memory exhaustion\n\n";
|
||||
|
||||
// Then try our new streaming approach
|
||||
echo "=== NEW STREAMING APPROACH ===\n";
|
||||
testNewApproach($outdatedPagesFile);
|
||||
|
||||
echo "Done testing!\n";
|
||||
Loading…
Add table
Add a link
Reference in a new issue