Finish improvements to pageindex rebuilder

also squash warning from stats engine during the firstrun wizard

Changelog.md

@@ -7,6 +7,7 @@ Make sure you have PHP 7.3+ when you update past this point! It isn't the end of
 
 ### Added
  - [Module Api] Add new `search::invindex_term_getpageids`, and `search::invindex_term_getoffsets`, and `search::index_sort_freq` methods
+ - [Module Api] Add new `ends_with` and `filepath_to_pagename` core functions
  - Added new syntax features to PeppermintParsedown, inspired by ParsedownExtreme (which we couldn't get to work, and it wasn't working before as far as I can tell)
      - Checkboxes: `[ ]` and `[x]` after a bullet point or at the start of a line
      - Marked / highlighted text: `Some text ==marked text== more text`
@@ -42,6 +43,7 @@ Make sure you have PHP 7.3+ when you update past this point! It isn't the end of
      - A warning is generated in PHP 7.2 and below = [please upgrade](https://www.php.net/supported-versions.php) to PHP 7.3+! (#200)
  - [security] The `Secure` cookie flag is now automatically added when clients use HTTPS to prevent downgrade-based session stealing attacks (control this with the new `cookie_secure` setting)
  - Standardised prefixes to (most) `error_log()` calls to aid clarity in multi-wiki environments
+ - Improved pageindex rebuilder algorithm to search for and import history revisions - this helps when converting data from another wiki format
 
 ### Fixed
  - Squashed a warning when using the fenced code block syntax
@@ -58,6 +60,7 @@ Make sure you have PHP 7.3+ when you update past this point! It isn't the end of
  - Fixed an obscure warning when previewing PDFs (#202)
  - Ensure that the parent page exists when moving a page to be a child of a non-existent parent (#201)
  - Fixed templating (#203)
+ - Fixed warning from statistics engine during firstrun wizard
 
 
 ## v0.21.1-hotfix1

core/05-functions.php

@@ -159,6 +159,30 @@ function path_resolve(string $path, string $basePath = null) {
 	return implode(DIRECTORY_SEPARATOR, $components);
 }
 
+/**
+ * Converts a filepath to a page name.
+ * @param  string $filepath The filepath to convert.
+ * @return string           The extracted pagename.
+ */
+function filepath_to_pagename(string $filepath) : string {
+	global $env;
+	// Strip the storage prefix, but only if it isn't a dot
+	if(starts_with($filepath, $env->storage_prefix) && $env->storage_prefix !== ".") {
+		$filepath = substr($filepath, strlen($env->storage_prefix));
+		// Strip the forward slash at the beginning
+		if($filepath[0] == "/" && $env->storage_prefix[-1] !== "/")
+			$filepath = substr($filepath, 1);
+	}
+	
+	if(preg_match("/\.r[0-9]+$/", $filepath) !== false)
+		$filepath = substr($filepath, 0, strrpos($filepath, ".r"));
+	
+	if(ends_with($filepath, ".md"))
+		$filepath = substr($filepath, 0, -3);
+	
+	return $filepath;
+}
+
 /**
  * Gets the name of the parent page to the specified page.
  * @apiVersion 0.15.0
@@ -303,10 +327,22 @@ function hide_email($str)
  *                        		of $haystack.
  * @return	bool	Whether $needle can be found at the beginning of $haystack.
  */
-function starts_with($haystack, $needle) {
+function starts_with(string $haystack, string $needle) : bool {
 	$length = strlen($needle);
 	return (substr($haystack, 0, $length) === $needle);
 }
+/**
+ * Checks to see if $hackstack ends with $needle.
+ * The matching bookend to starts_with.
+ * @package	core
+ * @param	string	$haystack	The haystack to search..
+ * @param	string	$needle		The needle to look for.
+ * @return	bool
+ */
+function ends_with(string $haystack, string $needle) : bool {
+	$length = strlen($needle);
+	return (substr($haystack, -$length) === $needle);
+}
 
 /**
  * Case-insensitively finds all occurrences of $needle in $haystack. Handles

core/20-pageindex-loader.php

@@ -103,16 +103,45 @@ if(!file_exists($paths->pageindex))
 	
 	if(function_exists("history_add_revision")) {
 		$history_revs = glob_recursive($env->storage_prefix . "*.r*");
+		// It's very important that we read the history revisions in the right order and that we don't skip any
+		usort($history_revs, function($a, $b) {
+			preg_match("/[0-9]+$/", $a, $revid_a);
+			$revid_a = intval($revid_a[0]);
+			preg_match("/[0-9]+$/", $b, $revid_b);
+			$revid_b = intval($revid_b[0]);
+			return $revid_a - $revid_b;
+		});
+		$strlen_storageprefix = strlen($env->storage_prefix);
 		foreach($history_revs as $filename) {
-			preg_match("/[0-9]+$/", "Main Page.md.r0", $revid);
+			preg_match("/[0-9]+$/", $filename, $revid);
+			error_log("raw revid | ".var_export($revid, true));
 			if(count($revid) === 0) continue;
 			$revid = intval($revid[0]);
 			
-			// TODO: Extract the pagename here (maybe a function is worth implementing if we haven't already?)
+			$pagename = filepath_to_pagename($filename);
+			$filepath_stripped = substr($filename, $strlen_storageprefix);
 			
-			if($revid == 0 && ) {
+			if(!isset($pageindex->$pagename->history))
+				$pageindex->$pagename->history = [];
 			
+			if(isset($pageindex->$pagename->history[$revid]))
+				continue;
+			
+			error_log("pagename: $pagename, revid: $revid, pageindex entry: ".var_export($pageindex->$pagename, true));
+			$newsize = filesize($filename);
+			$prevsize = 0;
+			if($revid > 0 && isset($pageindex->$pagename->history[$revid - 1])) {
+				$prevsize = filesize(end($pageindex->$pagename->history)->filename);
 			}
+			$pageindex->$pagename->history[$revid] = (object) [
+				"type" => "edit",
+				"rid" => $revid,
+				"timestamp" => filemtime($filename),
+				"filename" => $filepath_stripped,
+				"newsize" => $newsize,
+				"sizediff" => $newsize - $prevsize,
+				"editor" => "unknown"
+			];
 		}
 	}
 	

modules/feature-stats.php

@@ -1,7 +1,7 @@
 <?php
 register_module([
 	"name" => "Statistics",
-	"version" => "0.4.1",
+	"version" => "0.4.2",
 	"author" => "Starbeamrainbowlabs",
 	"description" => "An extensible statistics calculation system. Comes with a range of built-in statistics, but can be extended by other modules too.",
 	"id" => "feature-stats",
@@ -302,6 +302,10 @@ function update_statistics($update_all = false, $force = false)
 {
 	global $settings, $env, $paths, $statistic_calculators;
 	
+	// If the firstrun wizard isn't complete, then there's no point in updating the statistics index
+	if(isset($settings->firstrun_complete) && $settings->firstrun_complete == false)
+		return;
+	
 	$stats_mtime = filemtime($paths->statsindex);
 	
 	// Clear the existing statistics if we are asked to recalculate them all