SPIKE: highlight withdrawals

ynr/apps/elections/templates/elections/sopn_for_ballot.html

@@ -67,6 +67,7 @@ <h2>
             {% include "elections/includes/_sopn_debug.html" %}
         {% endif %}
 
+
         <div id="sopn-{{ object.ballot_paper_id }}" class="pdf_container"></div>
 
     {% else %}
@@ -78,13 +79,14 @@ <h2>
         {% url 'admin:official_documents_ballotsopn_change' object.sopn.id as url %}
         You can <a href="{{ url }}">edit this in the admin interface</a> (e.g. to delete it)
     {% endif %}
-
     {% if object.sopn.uploaded_file.url|slice:"-3:" == "pdf" %}
         <script type="module">
-            import { SOPN_VIEWER } from '/upload_document/sopn_viewer.js';
+            import {SOPN_VIEWER} from '/upload_document/sopn_viewer.js';
+
             SOPN_VIEWER.ShowSOPNInline(
                 '{{ object.sopn.uploaded_file.url }}',
-                '{{ object.ballot_paper_id }}'
+                '{{ object.ballot_paper_id }}',
+                {{ object.sopn.awstextractparsedsopn.get_withdrawals_bboxes|safe }}
             )
         </script>
     {% else %}

ynr/apps/official_documents/templates/official_documents/sopn_viewer.js

@@ -9,7 +9,17 @@ var SOPN_VIEWER = (function () {
 
     var module = {};
 
-    function load_page(pdf, container, page_num) {
+    function drawRectangles(context, rectangles) {
+        rectangles.forEach(rect => {
+            context.beginPath();
+            context.rect(rect.x, rect.y, rect.width, rect.height);
+            context.lineWidth = rect.lineWidth || 1;
+            context.strokeStyle = rect.color || 'red';
+            context.stroke();
+        });
+    }
+
+    function load_page(pdf, container, page_num, rectanglesPerPage) {
         return pdf.getPage(page_num).then(function (page) {
 
             var scale = 1.2;
@@ -25,6 +35,8 @@ var SOPN_VIEWER = (function () {
                 var context = canvas.getContext("2d");
                 canvas.height = viewport.height;
                 canvas.width = viewport.width;
+                console.log(canvas.height)
+                console.log(canvas.width)
                 var renderContext = {
                     canvasContext: context,
                     viewport: viewport
@@ -33,6 +45,12 @@ var SOPN_VIEWER = (function () {
                 var renderTask = page.render(renderContext);
                 return renderTask.promise.then(function () {
                     container.append(page_container);
+
+                    if (rectanglesPerPage && rectanglesPerPage[page_num]) {
+                        drawRectangles(context, rectanglesPerPage[page_num]);
+                    }
+
+
                     return page.getTextContent({normalizeWhitespace: true});
                 }).then(function (textContent) {
                     var pdf_canvas = $(canvas),
@@ -57,14 +75,16 @@ var SOPN_VIEWER = (function () {
                         viewport: viewport,
                         textDivs: []
                     });
+
+
                 });
 
             }
 
         });
     }
 
-    function ShowSOPNInline(sopn_url, ballot_paper_id, options) {
+    function ShowSOPNInline(sopn_url, ballot_paper_id, rectanglesPerPage) {
         // The container element
         var this_pdf_container = document.getElementById("sopn-" + ballot_paper_id);
 
@@ -73,7 +93,7 @@ var SOPN_VIEWER = (function () {
         loadingTask.promise.then(function (pdf) {
             var promise = Promise.resolve();
             for (let page = 1; page <= pdf.numPages; page++) {
-                promise = promise.then(() => load_page(pdf, this_pdf_container, page));
+                promise = promise.then(() => load_page(pdf, this_pdf_container, page, rectanglesPerPage));
             }
             return promise;
         }).then(null, function (error) {

ynr/apps/sopn_parsing/models.py

@@ -140,3 +140,45 @@ def parse_raw_data(self):
 
     def as_textractor_document(self):
         return response_parser.parse(json.loads(self.raw_data))
+
+    def get_withdrawals_bboxes(self):
+        # headers = self.as_pandas.iloc[0].tolist()
+        # get colmun index from headers
+        column = "5"
+        column_values = self.as_pandas[column].tolist()
+        cells_with_value = []
+        for i, row in enumerate(column_values):
+            if row:
+                cells_with_value.append(i)
+        cells_with_value.pop(0)
+        # Deal with more than one page
+        textract_cells = []
+        for table in self.as_textractor_document().tables:
+            for cell in table.table_cells:
+                # if str(cell.col_index-1) != column:
+                #     continue
+                if cell.row_index - 1 in cells_with_value:
+                    textract_cells.append(cell)
+        print(textract_cells)
+
+        doc_height = 1429
+        doc_width = 1010
+
+        page = 1
+        box_data = {page: []}
+        for cell in textract_cells:
+            absolute_x = cell.x * doc_width
+            absolute_y = cell.y * doc_height
+            absolute_width = cell.width * doc_width
+            absolute_height = cell.height * doc_height
+            box_data[page].append(
+                {
+                    "x": absolute_x,
+                    "y": absolute_y,
+                    "width": absolute_width,
+                    "height": absolute_height,
+                    "color": "red",
+                    "lineWidth": 2,
+                },
+            )
+        return json.dumps(box_data)