also re-run html modifyer

Dockerfile

@@ -42,7 +42,7 @@ COPY . /data/example/
 # move executables to PATH
 RUN mv /data/example/utils/watcher_docker.sh /usr/local/bin/watcher.sh
 RUN mv /data/example/utils/entrypoint.sh /usr/local/bin/cowkit.sh
-RUN mv /data/example/utils/html_parser.js /usr/local/bin/html_parser.js
+RUN mv /data/example/utils/html_* /usr/local/bin/
 
 ## install other latex engine
 #RUN curl --proto '=https' --tlsv1.2 -fsSL https://drop-sh.fullyjustified.net | sh

src/scripts/subfigs.js

@@ -19,8 +19,13 @@ function addSubfigCaption() {
       figcap.textContent = `(${img.alt})`;
       figure.style.width = img.style.width;
       img.style.width = "100%";
-      subfig.insertBefore(figure, img);
-      figure.appendChild(img);
+      let oldNode = img;
+      // use parentNode on img in case of "hyperlinked" img
+      if (img.parentNode.tagName == "A"){
+        oldNode = img.parentNode
+      }
+      subfig.insertBefore(figure, oldNode);
+      figure.appendChild(oldNode);
       figure.appendChild(figcap);
     })
   });

utils/entrypoint.sh

@@ -197,11 +197,15 @@ run_pandoc () {
         cp -rf $SCRIPTS_DIR $html_chunkdir
 
         # create search index
-        # html_parsers <file-in> <file-out> <jstemplate>
-        node /usr/local/bin/html_parser.js \
+        # html_search_index <file-in> <file-out> <jstemplate>
+        node /usr/local/bin/html_search_index.js \
             "$html_chunkdir/index.html" \
             "$ASSETS_DIR/search.json"
 
+        # html_modify <file-in> <file-out> <jstemplate>
+        node /usr/local/bin/html_modify.js \
+            "$html_chunkdir/index.html"
+
         cp -rf "$ASSETS_DIR/search.json" $html_assetsdir/
 
         #find $html_chunkdir -name "*.html" | while read file; do
@@ -343,7 +347,11 @@ else
             cd /app
             echo "[INFO - main] launching file watcher..."
             # run ./utils/watcher_docker.sh in docker container
-            watcher.sh "$CONFIG_MAIN" "$CONFIG_FORMAT" "$TEMPLATE" "$servedir" "$HTTP_PORT"
+            watcher.sh "$CONFIG_MAIN" "$CONFIG_FORMAT" "$TEMPLATE" \
+                "$OUTPUT_DIR" $CHUNKED_HTML "$SUBDIR_CHUNKED_HTML" \
+                            "$ASSETS_DIR" "$CONFIG_MAIN_DIR" "$SCRIPTS_DIR" \
+                            "$servedir"
+
         fi
     fi
 fi

utils/html_modify.js

@@ -0,0 +1,115 @@
+// SPDX-FileCopyrightText: 2024 Senne Van Baelen
+//
+// SPDX-License-Identifier: Apache-2.0
+
+const jsdom = require("jsdom");
+const { JSDOM } = jsdom;
+var path = require('path');
+const fs = require('node:fs/promises');
+
+/********************************************************************/
+/* Config + Args                                                    */
+/********************************************************************/
+
+// read input
+var ARGS = process.argv.slice(2);
+
+var MAIN_FPATH = ARGS[0];
+var MAIN_DIRNAME = path.dirname(MAIN_FPATH);
+
+
+/********************************************************************/
+/* Helpers                                                          */
+/********************************************************************/
+
+function escapeId(id) {
+  return id.replace(':', '\\:')
+}
+
+async function readFile(fpath) {
+  try {
+    const data = await fs.readFile(fpath, { encoding: 'utf8' });
+    return data;
+  } catch (err) {
+    console.log(err);
+  }
+}
+
+async function clickable_images(sHTML) {
+
+  const dom = new JSDOM(sHTML);
+  const imgElems = dom.window.document.querySelectorAll(`img`);
+
+  imgElems.forEach(img => {
+    const link = dom.window.document.createElement('a');
+    link.innerHTML = img.outerHTML;
+    //link.setAttribute('href', img.href);
+    link.setAttribute('href', img.src);
+    link.setAttribute('target', "_blank");
+    img.parentNode.insertBefore(link, img);
+    img.remove()
+  });
+
+  return dom;
+}
+
+async function process_chuncked_html(sHTML) {
+
+  const dom = new JSDOM(sHTML);
+
+  // The script will be executed and modify the DOM:
+  var tocContent = dom.window.document.querySelector("#toc-content");
+
+  var tocLinkArr = [];
+  var links = tocContent.getElementsByTagName("a");
+  for(var i=0; i<links.length; i++) {
+    tocLinkArr.push(links[i].href);
+  }
+
+  //console.log(tocLinkArr);
+  var fileContent = null;
+  var prevFileName = null;
+  var fileDom = null;
+  var arrFileObj = {};
+
+  for (const [idx, link] of tocLinkArr.entries()) {
+    const fileIdSplit = link.split('#');
+    const fileName = fileIdSplit[0];
+    const elId = escapeId(fileIdSplit[1]);
+    let dom = null;
+
+    if (fileName !== prevFileName){
+      fileContent = await readFile(`${MAIN_DIRNAME}/${fileName}`);
+      prevFileName = fileName;
+      dom = await clickable_images(fileContent);
+
+      const res = dom.serialize();
+      if (dom){
+        await fs.writeFile(`${MAIN_DIRNAME}/${fileName}`, res,
+          'utf8', function(err) {
+            if (err) throw err;
+          });
+      }
+    }
+
+  }
+
+}
+
+
+/********************************************************************/
+/* Main                                                             */
+/********************************************************************/
+
+async function main() {
+
+  var mainFileContent = await readFile(MAIN_FPATH);
+  //var sDoc = await generate_search_documents(arrFileObj);
+  var arrFileObj = await process_chuncked_html(mainFileContent);
+
+  /*await fs.writeFile(FPATH_JSON_OUT, json, 'utf8', function(err) {*/
+  /*if (err) throw err;*/
+  /*});*/
+}
+
+main();

utils/watcher_docker.sh

@@ -40,22 +40,76 @@ CONFIG_MAIN=$1
 CONFIG_FORMAT=$2
 TEMPLATE=$3
 OUTPUT_DIR=$4
+CHUNKED_HTML=$5
+SUBDIR_CHUNKED_HTML=$6
+ASSETS_DIR=$7
+CONFIG_MAIN_DIR=$8
+SCRIPTS_DIR=$9
+SERVER_DIR=${10}
 
 #=========================================================
 # Functions
 #=========================================================
 
 run_pandoc () {
 
-    #rm -rf "$OUTPUT_DIR"
-
     pandoc \
       --defaults=${CONFIG_MAIN} \
       --defaults=${CONFIG_FORMAT} \
       --template=${TEMPLATE} \
-      --output="${OUTPUT_DIR}_tmp"
+      --output="${SERVER_DIR}_tmp"
+
+    # html_modify <file-in> <file-out> <jstemplate>
+    node /usr/local/bin/html_modify.js \
+            "${SERVER_DIR}_tmp/index.html"
 }
 
+#run_pandoc () {
+    #config_format=$1
+    #template=$2
+
+    #if [ $CHUNKED_HTML = 1 ]; then
+        #html_chunkdir="$OUTPUT_DIR/$SUBDIR_CHUNKED_HTML"
+        #echo $html_chunkdir
+        ## make sure it's empty
+        #rm -rf $html_chunkdir
+    #fi
+
+    #pandoc \
+        #--defaults=${CONFIG_MAIN} \
+        #--defaults=${config_format} \
+        #--template=${template}
+
+    ## check if unzipping chunked HTML is necessary
+    ## unzipping is not necessary as it can be done through pandoc too, however,
+    ## the copying still is
+    #if [ $CHUNKED_HTML = 1 ]; then
+
+        ## unzip if chunked html
+        #html_chunkdir="$OUTPUT_DIR/$SUBDIR_CHUNKED_HTML"
+        #htmlcnfdir="$html_chunkdir/$CONFIG_MAIN_DIR/html"
+        #html_assetsdir="$html_chunkdir/assets"
+        #mkdir -p $htmlcnfdir
+        #mkdir -p $html_assetsdir
+
+        #cp -rf $CONFIG_MAIN_DIR/html/*  $htmlcnfdir
+        #cp -rf $SCRIPTS_DIR $html_chunkdir
+
+        ## create search index
+        ## html_search_index <file-in> <file-out> <jstemplate>
+        #node /usr/local/bin/html_search_index.js \
+            #"$html_chunkdir/index.html" \
+            #"$ASSETS_DIR/search.json"
+
+        ## html_modify <file-in> <file-out> <jstemplate>
+        #node /usr/local/bin/html_modify.js \
+            #"$html_chunkdir/index.html"
+
+        #cp -rf "$ASSETS_DIR/search.json" $html_assetsdir/
+    #fi
+#}
+
+
 #=========================================================
 # Main
 #=========================================================
@@ -102,12 +156,14 @@ do
 
         LAST_EVENT_TIME=$EVENT_TIME
         echo "[INFO - watcher] running pandoc"
-        run_pandoc
+        #run_pandoc
+        run_pandoc $CONFIG_FORMAT $TEMPLATE
+
         echo "[INFO - watcher] done - build output updated"
 
         # next two lines are to avoid pandoc error "createDirectory" failed (alreayd exist)
-        rsync --remove-source-files  --include "*.html" -a ${OUTPUT_DIR}_tmp/* $OUTPUT_DIR/
-        rm -rf ${OUTPUT_DIR}_tmp
+        rsync --remove-source-files  --include "*.html" -a ${SERVER_DIR}_tmp/* $SERVER_DIR/
+        rm -rf ${SERVER_DIR}_tmp
 
         # reload
         # first arg, or default to firefox if not set
@@ -123,9 +179,10 @@ do
             sleep $TIME_DIFF
             echo "[INFO - watcher] running pandoc"
             # next two lines are to avoid pandoc error "createDirectory" failed (alreayd exist)
-            run_pandoc
-            rsync --remove-source-files  --include "*.html" -a ${OUTPUT_DIR}_tmp/* $OUTPUT_DIR/
-            rm -rf ${OUTPUT_DIR}_tmp
+            #run_pandoc
+            run_pandoc $CONFIG_FORMAT $TEMPLATE
+            rsync --remove-source-files  --include "*.html" -a ${SERVER_DIR}_tmp/* $SERVER_DIR/
+            rm -rf ${SERVER_DIR}_tmp
         else
             echo "[INFO - watcher] skipping rerun (timediff set to $TIME_UNTIL_RERUN sec)"
         fi