From 4d44eb61841771a5d5d0032a45a4dd70e862e197 Mon Sep 17 00:00:00 2001
From: notmasteryet <async.processingjs@yahoo.com>
Date: Sun, 11 Dec 2011 18:14:52 -0600
Subject: [PATCH] fix stream reset, interrupting thread when pages are indexed

---
 src/core.js   |  4 ++--
 src/worker.js | 23 +++++++++++++++++------
 2 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/src/core.js b/src/core.js
index 6a90357e8..6a932f127 100644
--- a/src/core.js
+++ b/src/core.js
@@ -205,7 +205,7 @@ var Page = (function PageClosure() {
           streams.push(xref.fetchIfRef(content[i]));
         content = new StreamsSequenceStream(streams);
       } else if (isStream(content))
-        content.pos = 0;
+        content.reset();
 
       var pe = this.pe = new PartialEvaluator(
                                 xref, handler, 'p' + this.pageNumber + '_');
@@ -236,7 +236,7 @@ var Page = (function PageClosure() {
           streams.push(xref.fetchIfRef(content[i]));
         content = new StreamsSequenceStream(streams);
       } else if (isStream(content))
-        content.pos = 0;
+        content.reset();
 
       var pe = new PartialEvaluator(
                      xref, handler, 'p' + this.pageNumber + '_');
diff --git a/src/worker.js b/src/worker.js
index c3992e54f..3cc91d07e 100644
--- a/src/worker.js
+++ b/src/worker.js
@@ -164,23 +164,34 @@ var WorkerMessageHandler = {
     handler.on('extract_text', function wphExtractText() {
       var numPages = pdfDoc.numPages;
       var index = [];
-      for (var i = 0; i < numPages; i++) {
-        var start = Date.now();
+      var start = Date.now();
+
+      function indexPage(pageNum) {
+        if (pageNum > numPages) {
+          console.log('text indexing=: time=%dms', Date.now() - start);
+
+          handler.send('text_extracted', { index: index });
+          return;
+        }
 
         var textContent = '';
         try {
-          var page = pdfDoc.getPage(i + 1);
+          var page = pdfDoc.getPage(pageNum);
           textContent = page.extractTextContent();
         } catch (e) {
           // Skip errored pages
         }
 
         index.push(textContent);
+
+        // processing one page, interrupting thread to process
+        // other requests
+        setTimeout(function extractTextNextPage() {
+          indexPage(pageNum + 1);
+        }, 0);
       }
 
-      console.log('text indexing=: time=%dms', Date.now() - start);
-
-      handler.send('text_extracted', { index: index });
+      indexPage(1);
     });
   }
 };