Default toc length config

dev/config/pyramid_oereb.yml.mako

@@ -88,6 +88,16 @@ pyramid_oereb:
     # more time to generate the PDF. If set to false, it will assume that only one TOC page exists, and this can
     # lead to wrong numbering in the TOC.
     compute_toc_pages: true
+    # To avoid the potentially time consuming second computing of the PDF extract and skip the the computation
+    # of the estimated TOC length, you can specify a default length for the number of TOC pages.
+    # For most of the cantons the length of the TOC is pretty consistent unless a real estate is concerned by none 
+    # or a huge number of restrictions.
+    # An additional page break might also occur if the number of published topics is close to a threshold number
+    # where the TOC fits just about on one or two pages. - for those case estimate the TOC length ist preferable.
+    # In both cases (computing an estimated length or setting a default length) the exact number of TOC pages is
+    # extracted from the created PDF and if it is different from the expected value the PDF extract is called a 
+    # second time with the correct page numbers.
+    default_toc_length: 2
     # Specify any additional URL parameters that the print shall use for WMS calls
     wms_url_params:
       TRANSPARENT: 'true'

doc/source/changes.rst

@@ -6,6 +6,13 @@ Changes/Hints for migration
 This chapter will give you hints on how to handle version migration, in particular regarding what you may need
 to adapt in your project configuration, database etc. when upgrading to a new version.
 
+Version 2.6.0
+-------------
+* New parameter 'default_toc_length' allows to define a default table of content pages number avoiding a second
+call for the pdf extract in most cases. This value should be set if >95% of the PDF have the same number of TOC 
+pages.
+Default setting: 2
+
 Version 2.5.3
 -------------
 Feature and maintenance release:

pyramid_oereb/contrib/print_proxy/mapfish_print/mapfish_print.py

@@ -73,17 +73,20 @@
         extract_as_dict = self._render(extract_record, value[1])
         feature_geometry = mapping(extract_record.real_estate.limit)
 
+        print_config = Config.get('print', {})
+
         if Config.get('print', {}).get('compute_toc_pages', False):
             extract_as_dict['nbTocPages'] = TocPages(extract_as_dict).getNbPages()
         else:
-            extract_as_dict['nbTocPages'] = 1
+            if Config.get('print', {}).get('default_toc_length', False):
+                extract_as_dict['nbTocPages'] = print_config.get('default_toc_length', 2)
+            else:
+                extract_as_dict['nbTocPages'] = 1
 
         # set the global_datetime variable so that it can be used later for the archive
         self.set_global_datetime(extract_as_dict['CreationDate'])
         self.convert_to_printable_extract(extract_as_dict, feature_geometry)
 
-        print_config = Config.get('print', {})
-
         extract_as_dict['Display_RealEstate_SubunitOfLandRegister'] = print_config.get(
             'display_real_estate_subunit_of_land_register', True
         )
@@ -116,6 +119,7 @@
             data=json.dumps(spec)
         )
         try:
+            log.debug('Validation of the TOC length with compute_toc_pages set to {} and default_toc_length set to {}'.format(print_config.get('compute_toc_pages'), print_config.get('default_toc_length'))) # noqa
             if Config.get('print', {}).get('compute_toc_pages', False):
                 with io.BytesIO() as pdf:
                     pdf.write(print_result.content)
@@ -131,14 +135,43 @@
                     except ValueError:
                         true_nb_of_toc = 1
 
+                    log.debug('True number of TOC pages is {}'.format(true_nb_of_toc))
+                    if true_nb_of_toc != extract_as_dict['nbTocPages']:
+                        log.warning('nbTocPages in result pdf: {} are not equal to the one predicted : {}, request new pdf'.format(true_nb_of_toc,extract_as_dict['nbTocPages'])) # noqa
+                        log.debug('Secondary PDF extract call STARTED')
+                        extract_as_dict['nbTocPages'] = true_nb_of_toc
+                        print_result = requests.post(
+                            pdf_url,
+                            headers=pdf_headers,
+                            data=json.dumps(spec)
+                        )
+                        log.debug('Secondary PDF extract call to fix TOC pages number FINISHED')
+            elif Config.get('print', {}).get('default_toc_length', 2):
+                with io.BytesIO() as pdf:
+                    pdf.write(print_result.content)
+                    pdf_reader = PdfReader(pdf)
+                    x = []
+                    for i in range(len(pdf_reader.outline)):
+                        if isinstance(pdf_reader.outline[i], list):
+                            x.append(pdf_reader.outline[i][0]['/Page']['/StructParents'])
+                        else:
+                            x.append(pdf_reader.outline[i]['/Page']['/StructParents'])
+                    try:
+                        true_nb_of_toc = min(x)-1
+                    except ValueError:
+                        true_nb_of_toc = 1
+
+                    log.debug('True number of TOC pages is {}'.format(true_nb_of_toc))
                     if true_nb_of_toc != extract_as_dict['nbTocPages']:
                         log.warning('nbTocPages in result pdf: {} are not equal to the one predicted : {}, request new pdf'.format(true_nb_of_toc,extract_as_dict['nbTocPages'])) # noqa
                         extract_as_dict['nbTocPages'] = true_nb_of_toc
+                        log.debug('Secondary PDF extract call STARTED')
                         print_result = requests.post(
                             pdf_url,
                             headers=pdf_headers,
                             data=json.dumps(spec)
                         )
+                        log.debug('Secondary PDF extract call FINISHED')
         except PdfReadError as e:
             err_msg = 'a problem occurred while generating the pdf file'
             log.error(err_msg + ': ' + str(e))

tests/contrib.print_proxy.mapfish_print/resources/test_config.yml

@@ -4,6 +4,8 @@ pyramid_oereb:
     wms_url_params:
       TRANSPARENT: 'true'
       OTHERCUSTOM: 'myvalue'
+    compute_toc_pages: false
+    default_toc_length: 2
 
   theme:
     source:

tests/contrib.print_proxy.mapfish_print/test_mapfish_print_configuration.py

@@ -100,3 +100,14 @@ def test_default_wms_url_param_config(DummyRenderInfo):
     config = renderer.get_wms_url_params()
     # Do the check for this test. Value should be the default setting.
     assert config == {'TRANSPARENT': 'true'}
+
+
+def test_toc_pages_default_config():
+    Config._config = None
+    Config.init('./tests/contrib.print_proxy.mapfish_print/resources/test_config.yml', 'pyramid_oereb')
+    compute_toc_pages = Config.get('print', {}).get('compute_toc_pages')
+    default_toc_length = Config.get('print', {}).get('default_toc_length')
+
+    assert isinstance(compute_toc_pages, bool)
+    assert bool(compute_toc_pages) is False
+    assert default_toc_length == 2