Fix table auto detection

camelot-dev · Nov 23, 2018 · 4f6b414 · 4f6b414
1 parent 4287ae9
commit 4f6b414
Show file tree

Hide file tree

Showing 4 changed files with 51 additions and 44 deletions.
diff --git a/excalibur/tasks.py b/excalibur/tasks.py
@@ -97,15 +97,15 @@ class GhostscriptNotFound(Exception): pass
             if len(tables):
                 lattice_areas = []
                 for table in tables:
-                    x1, y1, x2, y2 = tables[0]._bbox
+                    x1, y1, x2, y2 = table._bbox
                     lattice_areas.append((x1, y2, x2, y1))
             # stream
             parser = Stream()
             tables = parser.extract_tables(filepath)
             if len(tables):
                 stream_areas = []
                 for table in tables:
-                    x1, y1, x2, y2 = tables[0]._bbox
+                    x1, y1, x2, y2 = table._bbox
                     stream_areas.append((x1, y2, x2, y1))
 
             detected_areas[page] = {

diff --git a/excalibur/utils/task.py b/excalibur/utils/task.py
@@ -6,49 +6,49 @@
 
 
 def get_pages(filename, pages):
-        """Converts pages string to list of ints.
+    """Converts pages string to list of ints.
 
-        Parameters
-        ----------
-        filename : str
-            Path to PDF file.
-        pages : str, optional (default: '1')
-            Comma-separated page numbers.
-            Example: 1,3,4 or 1,4-end.
+    Parameters
+    ----------
+    filename : str
+        Path to PDF file.
+    pages : str, optional (default: '1')
+        Comma-separated page numbers.
+        Example: 1,3,4 or 1,4-end.
 
-        Returns
-        -------
-        N : int
-            Total pages.
-        P : list
-            List of int page numbers.
+    Returns
+    -------
+    N : int
+        Total pages.
+    P : list
+        List of int page numbers.
 
-        """
-        page_numbers = []
-        inputstream = open(filename, 'rb')
-        infile = PdfFileReader(inputstream, strict=False)
-        N = infile.getNumPages()
-        if pages == '1':
-            page_numbers.append({'start': 1, 'end': 1})
+    """
+    page_numbers = []
+    inputstream = open(filename, 'rb')
+    infile = PdfFileReader(inputstream, strict=False)
+    N = infile.getNumPages()
+    if pages == '1':
+        page_numbers.append({'start': 1, 'end': 1})
+    else:
+        if infile.isEncrypted:
+            infile.decrypt(self.password)
+        if pages == 'all':
+            page_numbers.append({'start': 1, 'end': infile.getNumPages()})
         else:
-            if infile.isEncrypted:
-                infile.decrypt(self.password)
-            if pages == 'all':
-                page_numbers.append({'start': 1, 'end': infile.getNumPages()})
-            else:
-                for r in pages.split(','):
-                    if '-' in r:
-                        a, b = r.split('-')
-                        if b == 'end':
-                            b = infile.getNumPages()
-                        page_numbers.append({'start': int(a), 'end': int(b)})
-                    else:
-                        page_numbers.append({'start': int(r), 'end': int(r)})
-        inputstream.close()
-        P = []
-        for p in page_numbers:
-            P.extend(range(p['start'], p['end'] + 1))
-        return sorted(set(P)), N
+            for r in pages.split(','):
+                if '-' in r:
+                    a, b = r.split('-')
+                    if b == 'end':
+                        b = infile.getNumPages()
+                    page_numbers.append({'start': int(a), 'end': int(b)})
+                else:
+                    page_numbers.append({'start': int(r), 'end': int(r)})
+    inputstream.close()
+    P = []
+    for p in page_numbers:
+        P.extend(range(p['start'], p['end'] + 1))
+    return sorted(set(P)), N
 
 
 def save_page(filepath, page_number):

diff --git a/excalibur/www/static/js/workspace.js b/excalibur/www/static/js/workspace.js
@@ -229,9 +229,16 @@ const renderTableAreas = function (page, tableAreas) {
 };
 
 const onDetectAreasClick = (e) => {
-  const flavor = e.getAttribute('data-flavor');
   for (let page in detectedAreas) {
+    let flavor = '';
+    if (detectedAreas[page]['lattice'] != null) {
+      flavor = 'lattice';
+    } else {
+      flavor = 'stream';
+    }
     renderTableAreas(page, detectedAreas[page][flavor]);
+    onFlavorChange();
+    document.getElementById('flavors').value = flavor.charAt(0).toUpperCase() + flavor.slice(1);
   }
 }
 

diff --git a/excalibur/www/templates/workspace.html b/excalibur/www/templates/workspace.html
@@ -31,8 +31,8 @@ <h6 class="dropdown-header">Saved Rules</h6>
                 {% endif %}
               </div>
             </div>
-            <button type="button" class="btn btn-secondary detect-areas mr-2" data-flavor="lattice" onclick="onDetectAreasClick(this)"><i class="fas fa-eye mr-2"></i>Autodetect Tables</button>
-            <button type="button" id="reset-areas" class="btn btn-secondary reset-areas mr-2" onclick="resetTableAreas()"><i class="fas fa-trash-alt mr-2"></i>Clear Selections</button>
+            <button type="button" class="btn btn-secondary detect-areas mr-2" onclick="onDetectAreasClick(this)"><i class="fas fa-eye mr-2"></i>Autodetect Tables</button>
+            <button type="button" id="reset-areas" class="btn btn-secondary reset-areas mr-2" onclick="resetTableAreas()"><i class="fas fa-trash-alt mr-2"></i>Clear Tables</button>
           </div>
         </section>
         <section class="col-md-2">