Merge pull request #2 from jxmot/20220611-refactor

Merge refactoring changes

.gitignore

@@ -0,0 +1 @@
+*.sh

README.md

@@ -1,6 +1,6 @@
 # minimize-prep
 
-This repository contains a PHP script that reads a file where HTML tags (`<script>` and `<link>`) are used for including CSS and Javascript files in a web page. As each CSS or JavaScript file tag is found the file contents are read and then appended to an appropriate output file (CSS or JS).
+This repository contains a PHP script that reads a file where HTML tags (`<script>` and `<link>`) are used for including CSS and Javascript files in a web page. As each CSS or JavaScript file tag is found the file contents are read and then appended to an appropriate output file (CSS or JS). 
 
 In any `<script>` and `<link>` tags where the resource is not local those will be skipped and their contents will not be appended to output files (CSS or JS). 
 
@@ -10,6 +10,7 @@ The result are two files that are ready for *minimization*. However, minimizatio
 
 * `minprep.json` - script configuration
 * `minprep.php` - opens the specified (*in minprep.json*) HTML file, reads it and looks for `<link>` and `<script>` tags. When found the path and file name are extracted and those files will be opened and concatenated in the specified CSS and JS files.
+* `minimize-prep.php` - Contains the functions used by `minprep.php`.
 * `public_html/*` - this folder contains files for use in the example
 
 ## Set Up
@@ -58,10 +59,10 @@ Now open `minprep.json`:
     "fileroot":"./public_html/",
     "cssexclude": ["cssexcl"],
     "jsexclude": ["jsexcl"],
-    "cssout": "./site.css",
-    "jsout": "./site.js",
-    "filecomment": true,
-    "mkbash": true
+    "cssout": "site.css",
+    "jsout": "site.js",
+    "mkremove": true,
+    "rmvscript": "rmvresources.sh"
 }
 ```
 
@@ -71,8 +72,7 @@ The remaining settings are:
 
 * `"cssexclude": ["cssexcl"]` and `"jsexclude": ["jsexcl"]` - Each is an array of strings where each is compared against the current resource found in `./public_html/example.html`. If there is a *partial* match then that resource will be excluded from the `./site.css` or `./site.js` files.
 * `"cssout": "./site.css"` and `"jsout": "./site.js"` - The path + file names of the resulting output files. 
-* `"filecomment": true` - For convenience, when `true` a commented line with the current resource file name will precede its contents. 
-* `"mkbash": true` - When `true` a file named `./rmvresources.sh` will be created. It will contain Linux `rm` commands for each of the resource files found.
+* `"mkremove": true` - When `true` a file named in `"rmvscript"` will be created. It will contain Linux `rm` commands for each of the resource files found.
 
 # Run
 
@@ -86,29 +86,23 @@ The script will create output similar to this:
 Starting preparation...
 Input: ./public_html/example.html
 Files Root Path: ./public_html/
-./site.css and ./site.js will be overwritten.
-
-Creating ./rmvresources.sh file
-
-CSS found - ./assets/css/example_1.css
-CSS found - ./assets/css/example_2.css
-CSS found - assets/css/example_3.css
-
-CSS exluded -      <link rel="stylesheet" href="./assets/css/cssexclude_1.css" type="text/css" />
-
-
-CSS exluded -      <link rel="stylesheet" href="./assets/cssexcl/exclude_1.css" type="text/css" />
-
-JS found - /assets/js/example_1.js
-JS found - ./assets/js/example_2.js
-JS found - ./assets/js/example_3.js
-JS found - ./assets/js/example_4.js
-
-JS exluded -      <script src="/assets/js/jsexclude_1.js" type="text/javascript"></script>
-
-
-JS exluded -      <script src="/assets/jsexcl/exclude_1.js" type="text/javascript"></script>
-
+./public_html/assets/css/site.css and ./public_html/assets/js/site.js will be overwritten.
+
+Creating rmvresources.sh file
+
+Excluded - <link rel="stylesheet" href="https://unpkg.com/[email protected]/reset.css" />
+Found - ./assets/css/example_1.css
+Found - ./assets/css/example_2.css
+Found - assets/css/example_3.css
+Excluded - <link rel="stylesheet" href="./assets/css/cssexclude_1.css" type="text/css" />
+Excluded - <link rel="stylesheet" href="./assets/cssexcl/exclude_1.css" type="text/css" />
+Excluded - <script src="//code.jquery.com/jquery-3.6.0.min.js" type="text/javascript"></script>
+Found - /assets/js/example_1.js
+Found - ./assets/js/example_2.js
+Found - ./assets/js/example_3.js
+Found - ./assets/js/example_4.js
+Excluded - <script src="/assets/js/jsexclude_1.js" type="text/javascript"></script>
+Excluded - <script src="/assets/jsexcl/exclude_1.js" type="text/javascript"></script>
 
 Preparation Complete.
 ```
@@ -120,7 +114,7 @@ The current configuration will cause 3 files to be created:
 * `site.css` and `site.js`
 * `rmvresources.sh` - An optional bash script file that will contain Linux `rm` commands to delete the CSS and JSS *source* files that were used to create `site.css` and `site.js`. **USE THIS SCRIPT WITH CAUTION!!!**
 
-To disable the creation of `rmvresources.sh` edit `minprep.json` and change `"mkbash"` to `false`.
+To disable the creation of `rmvresources.sh` edit `minprep.json` and change `"mkremove"` to `false`.
 
 # Important Things to Note
 
@@ -138,7 +132,7 @@ This will **not** work:
 -->
 ```
 
-All `<link>` and `<script>` tags need to be contained on a single line, for example this will not work:
+All `<link>` and `<script>` tags need to be contained on a single line, for example this will **not** work:
 
 ```
 <script
@@ -148,7 +142,7 @@ src="path/to/some.js"
 
 The paths in the `<link>` and `<script>` can be *relative* or *absolute*. The script will create a path using `"fileroot"`in the `minprep.json` file.
 
-Each time you run `minprep.php` it will **overwrite** the `site.css` and `site.js` files. And if `"mkbash"` is `true` in `minprep.json` then a *bash script* named `rmvresources.sh` will be created. Its purpose is to make removing the original CSS and JS easier. Although removing them is not required, it is your descision for your project.
+Each time you run `minprep.php` it will **overwrite** the `site.css` and `site.js` files. And if `"mkremove"` is `true` in `minprep.json` then a *bash script* named in `"rmvscript"`  will be created. Its purpose is to make removing the original CSS and JS easier. Although removing them is not required, it is your descision for your project.
 
 ---
 <img src="http://webexperiment.info/extcounter/mdcount.php?id=minimize-prep">

minimize-prep.php

@@ -0,0 +1,162 @@
+<?php
+/*
+    Does the line contain a <link>?
+
+    $hline      - a line of text read from the input file
+
+    Returns:
+        > 0     - it contains a <link>, return value is position of "href="
+        false   - it does not contain a <link>, or if it does then it's commented out 
+        -1      - it does not contain a <link>
+*/
+function isLink($hline) {
+    $ret = -1;
+    // is the line commented out? 
+    // get positions of markers...
+    $cbeg = strpos($hline, '<!--');
+    $cend = strpos($hline, '-->');
+    // first check...
+    if(strpos($hline, '<link') !== false) {
+        // second check...
+        if(strpos($hline, 'rel="stylesheet"') !== false) {
+            // get the href
+            $href = strpos($hline, 'href=');
+            // see if the "href=" is between comment markers
+            if($href === false || ($href > $cbeg) && ($href < $cend)) $ret = false;
+            else $ret = $href; // return position, > 0
+        }
+    }
+    return $ret;
+}
+
+/*
+    Does the line contain a <script>?
+
+    $hline      - a line of text read from the input file
+
+    Returns:
+        > 0     - it contains a <script>, return value is position of "src="
+        false   - it does not contain a <script>, or if it does then it's commented out 
+        -1      - it does not contain a <script>
+*/
+function isScript($hline) {
+    $ret = -1;
+    // is the line commented out? 
+    // get positions of markers...
+    $cbeg = strpos($hline, '<!--');
+    $cend = strpos($hline, '-->');
+
+    // first check...
+    // script?
+    if(strpos($hline, '<script') !== false) {
+        // get the src
+        $src = strpos($hline, 'src=');
+        // see if the "src=" is between comment markers
+        if($src === false || ($src > $cbeg) && ($src < $cend)) $ret = false;
+        else $ret = $src; // return position, > 0
+    }
+    return $ret;
+}
+
+/*
+    Determines if a line should be excluded 
+    from further processing.
+
+    $hline      - a line of text read from the input file
+    $exclude    - an array of strings, if found in $hline then exclude
+
+    Returns:
+        true    - exclude(skip) this line
+        false   - process this line
+*/
+function isExcluded($hline, $exclude) {
+    $excl = false;
+    // <link> or <script> containing the following 
+    // will be excluded
+    if((strpos($hline, 'http:') === false) && 
+       (strpos($hline, 'https:') === false) && 
+       (strpos($hline, 'site.css') === false) && 
+       (strpos($hline, 'site.js') === false) && 
+       (strpos($hline, 'site.min.css') === false) && 
+       (strpos($hline, 'site.min.js') === false) && 
+       (strpos($hline, 'jquery') === false) && 
+       (strpos($hline, '//') === false)) {
+        // check the exclusion list
+        if(count($exclude) > 0) {
+            for($ix = 0;$ix < count($exclude);$ix++) {
+                if(strpos($hline, $exclude[$ix]) !== false) {
+                    // exclude this one
+                    $excl = true;
+                    break;
+                }
+            }
+        }
+    } else {
+        $excl = true;
+    }
+    return $excl;
+}
+
+/*
+    Get the URL out of the line, starts at an 
+    offset and ends with a double-quote 
+
+    $hline      - a line of text read from the input file
+    $offset     - starting poistion where the URL begins
+
+    Returns:
+
+        A string containing the URL of the resource
+
+*/
+function getURL($hline, $offset) {
+    $url = substr($hline, $offset);
+    $url = substr($url, 0, strpos($url, '"'));
+    return $url;
+}
+
+/*
+    Get the full path to the resource
+
+    $url        - string returned from getURL()
+    $fileroot   - path to root of resource file tree
+
+    Returns:
+        Full tile path to the resource 
+*/
+function getFilePath($url, $fileroot) {
+    if(strpos($url, './') === 0) {
+        $fpath = $fileroot . substr($url, 2);
+    } else {
+        if(strpos($url, '/') === 0) {
+            $fpath = $fileroot . substr($url, 1);
+        } else {
+            $fpath = $fileroot . $url;
+        }
+    }
+    return $fpath;
+}
+
+/*
+    Put the contents of a resource into the 
+    output file.
+
+    $fpath      - the full path returned from getFilePath()
+    $fout       - file handle to the output file
+
+    Returns: 
+        true    - resource content was written to the output file 
+        false   - the resource file could not be found
+*/
+function putContents($fpath, $fout) {
+    if(file_exists($fpath) === false) {
+        return false;
+    } else {
+        $content = file_get_contents($fpath);
+        fwrite($fout, $content);
+        fwrite($fout, "\n");
+        fflush($fout);
+        return true;
+    }
+}
+?>

minprep.json

@@ -3,8 +3,8 @@
     "fileroot":"./public_html/",
     "cssexclude": ["cssexcl"],
     "jsexclude": ["jsexcl"],
-    "cssout": "site.css",
-    "jsout": "site.js",
-    "filecomment": true,
-    "mkbash": true
+    "cssout": "assets/css/site.css",
+    "jsout": "assets/js/site.js",
+    "mkremove": true,
+    "rmvscript": "rmvresources.sh"
 }