add pdf features (#23)

* add pdf features

* separatore csv

* black

* pdf descr

* black

* logo

* typo

* footer

* dt

* typo

* test

* remove file

* ignore

* rimosso supporto svg

* fix pdf css

* Preparing release 1.0.8

* Back to development: 1.0.9

---------

Co-authored-by: plone@umbar.hetzner.redturtle.it <plone@umbar.hetzner.redturtle.it>
Co-authored-by: Andrea Cecchi <andrea.cecchi85@gmail.com>

Author: 3 people

.gitignore

@@ -39,3 +39,5 @@ reports/
 venv/
 # excludes
 .python-version
+*.bak
+nohup.out

CHANGES.rst

@@ -2,7 +2,13 @@ Changelog
 =========
 
 
-1.0.8 (unreleased)
+1.0.9 (unreleased)
+------------------
+
+- Nothing changed yet.
+
+
+1.0.8 (2025-08-29)
 ------------------
 
 - Blocco search / variante table - aggiunte le proprietà dei campi nella colonna nel serializer

setup.py

@@ -16,7 +16,7 @@
 
 setup(
     name="iosanita.contenttypes",
-    version="1.0.8.dev0",
+    version="1.0.9.dev0",
     description="An add-on for Plone",
     long_description=long_description,
     # Get more from https://pypi.org/classifiers/

src/iosanita/contenttypes/browser/export_view.py

@@ -2,14 +2,18 @@
 from io import BytesIO
 from io import StringIO
 from iosanita.contenttypes import _
+from PIL import Image
 from plone import api
+from plone.memoize import forever
 from Products.Five.browser import BrowserView
 from weasyprint import HTML
-from zExceptions import BadRequest
+from zExceptions import NotFound
 from zope.interface import implementer
 from zope.publisher.interfaces import IPublishTraverse
 
+import base64
 import csv
+import imghdr
 import importlib.resources
 import logging
 import re
@@ -21,11 +25,58 @@
 fontools_logger.setLevel(logging.WARNING)
 
 
-CONTENT_TYPES_MAPPING = {
-    "csv": "text/comma-separated-values",
-    "pdf": "application/pdf",
-    "html": "text/html",
-}
+@forever.memoize
+def image_to_html(input_string):
+    """
+    Convert image data to a base64 string formatted for HTML.
+
+    Args:
+    - input_string: The string containing the filename and base64 encoded image data.
+
+    Returns:
+    - HTML.
+    """
+
+    if not input_string:
+        return ""
+
+    # Split the input string to extract the filename and base64 data
+    parts = input_string.split(";")
+    datab64 = parts[1].split(":")[1]
+
+    # Decode the image data from base64
+    image_data = base64.b64decode(datab64)
+
+    if image_data[:5] == b"<?xml":
+        # https://github.com/Kozea/WeasyPrint/issues/75
+        # anche se il ticket risulta chiuso gli svg non risultano correttamente gestiti
+        # return image_data
+        # return f'<img src="data:image/svg+xml;charset=utf-8;base64,{datab64}">'
+        # XXX: se non si va decode/encode il b64 non risulta corretto (!)
+        # return f'<img src="data:image/svg+xml;charset=utf-8;base64,{base64.b64encode(image_data).decode()}">'
+        # weasyprint gli svg non li gestisce comunque correttamente
+        return None
+
+    # Guess the image format
+    image_format = imghdr.what(None, image_data)
+
+    if not image_format:
+        # raise ValueError("Unable to determine image format")
+        logger.warning("site logo, unable to determine image format")
+        return ""
+
+    # Open the image from the decoded data
+    img = Image.open(BytesIO(image_data))
+
+    # Create a buffer to hold the image data
+    buffered = BytesIO()
+    img.save(buffered, format=image_format)
+
+    # Encode the image data to base64
+    img_base64 = base64.b64encode(buffered.getvalue()).decode("utf-8")
+
+    # Format the base64 string for HTML
+    return f'<img class="logo" src="data:{image_format};base64,{img_base64}">'
 
 
 class IExportViewTraverser(IPublishTraverse):
@@ -50,6 +101,8 @@ class ExportViewDownload(BrowserView):
 
     """
 
+    with_footer = True
+
     def __init__(self, context, request):
         super().__init__(context, request)
         self.export_type = "csv"
@@ -65,7 +118,7 @@ def publishTraverse(self, request, name):
     def __call__(self):
         """ """
         if self.export_type not in ["csv", "pdf", "html"]:
-            raise BadRequest(
+            raise NotFound(
                 api.portal.translate(
                     _(
                         "invalid_export_type",
@@ -74,18 +127,21 @@ def __call__(self):
                     )
                 )
             )
-        self.set_headers()
         data = self.get_data()
-        if not data:
-            return ""
-        resp_data = ""
         if self.export_type == "csv":
-            resp_data = self.get_csv(data)
+            # default per locales di riferimento (perr l'encoding, al momento, lasciamo
+            # il generico utf-8 con BOM che potrebbe funzionare per tutti,
+            # MS Excel incluso)
+            lang = api.portal.get_current_language(self.context)
+            if lang == "it":
+                sep = ";"
+            else:
+                sep = ","
+            return self.get_csv(data, sep=sep)
         elif self.export_type == "pdf":
-            resp_data = self.get_pdf(data)
+            return self.get_pdf(data)
         elif self.export_type == "html":
-            resp_data = self.get_html_for_pdf(data)
-        return resp_data
+            return self.get_html_for_pdf(data)
 
     def get_filename(self):
         """
@@ -94,37 +150,46 @@ def get_filename(self):
         now = datetime.now().strftime("%Y_%m_%d_%H_%M")
         return f"export_{now}.{self.export_type}"
 
-    def set_headers(self):
-        """
-        Set the headers for the response.
-        """
-        if self.export_type in ["pdf", "csv"]:
-            self.request.response.setHeader(
-                "Content-Disposition", f"attachment;filename={self.get_filename()}"
-            )
-        self.request.response.setHeader(
-            "Content-Type", CONTENT_TYPES_MAPPING[self.export_type]
-        )
-
-    def get_csv(self, data):
+    def get_csv(self, data, encoding="utf-8-sig", sep=","):
         """
         Generate CSV data from the provided data.
         """
+        # 1. Crea uno StringIO per il CSV
+        csv_buffer = StringIO()
+        # 2. Aggiungi l'header per il separatore (specifico per Excel)
+        # In Libreoffice viene aggiunta una riga, per ora evitiamo
+        # csv_buffer.write(f"sep={sep}\n")
+        # 3. Scrittura dei dati CSV
         columns = self.get_columns(data)
-
-        csv_data = StringIO()
-        csv_writer = csv.writer(csv_data, quoting=csv.QUOTE_ALL)
+        csv_writer = csv.writer(csv_buffer, delimiter=sep, quoting=csv.QUOTE_ALL)
         csv_writer.writerow([c["title"] for c in columns])
-
         for item in data:
             csv_writer.writerow(self.format_row(item))
-        return csv_data.getvalue().encode("utf-8")
+        # 4. Prepara i bytes con BOM (UTF-8-sig)
+        csv_data = csv_buffer.getvalue()
+        if encoding == "utf-8-sig":
+            csv_bytes = b"\xef\xbb\xbf" + csv_data.encode("utf-8")  # Aggiunge BOM
+        else:
+            csv_bytes = csv_data.encode(encoding)
+        # 5. Crea la risposta con gli header corretti
+        response = self.request.response
+        response.setHeader(
+            "Content-Disposition", f"attachment;filename={self.get_filename()}"
+        )
+        response.setHeader("Content-Type", f"text/csv; charset={encoding}")
+        return csv_bytes
 
     def get_pdf(self, data):
         html_str = self.get_html_for_pdf(data=data)
         pdf_file = BytesIO()
         HTML(string=html_str).write_pdf(pdf_file)
         pdf_file.seek(0)
+        # 5. Crea la risposta con gli header corretti
+        response = self.request.response
+        response.setHeader(
+            "Content-Disposition", f"attachment;filename={self.get_filename()}"
+        )
+        response.setHeader("Content-Type", "application/pdf")
         return pdf_file.read()
 
     def get_data(self):
@@ -176,6 +241,7 @@ def get_html_for_pdf(self, data):
             context=self,
             request=self.request,
         )
+
         return view(rows=data, columns=columns)
 
     def pdf_styles(self):
@@ -184,7 +250,11 @@ def pdf_styles(self):
         )
 
     def pdf_title(self):
-        return None
+        context = self.context.context
+        site_title = api.portal.get_registry_record("plone.site_title")
+        if site_title:
+            return f"{site_title}: {context.Title()}"
+        return context.Title()
 
     def pdf_description(self):
         return None
@@ -203,3 +273,13 @@ def pdf_cell_format(self, column, value):
         if re.match(r"^\d{4}-\d{2}-\d{2}T00:00:00$", value):
             return {"type": "str", "value": value.split("T")[0]}
         return {"type": "str", "value": str(value)}
+
+    def pdf_logo(self):
+        site_logo = api.portal.get_registry_record("plone.site_logo")
+        if site_logo:
+            return image_to_html(site_logo.decode())
+        return None
+
+    def pdf_datetime(self):
+        # TODO: valutare localizzazione della data
+        return datetime.now().strftime("%d/%m/%Y %H:%M")

src/iosanita/contenttypes/browser/searchblock.py

@@ -4,10 +4,15 @@
 from .export_view import IExportViewTraverser
 from copy import deepcopy
 from iosanita.contenttypes import _
+from plone.app.querystring.interfaces import IQuerystringRegistryReader
+from plone.intelligenttext.transforms import convertWebIntelligentPlainTextToHtml
+from plone.memoize import view
+from plone.registry.interfaces import IRegistry
 from plone.restapi.interfaces import ISerializeToJson
 from zExceptions import BadRequest
 from zExceptions import NotFound
 from zope.component import getMultiAdapter
+from zope.component import getUtility
 from zope.interface import implementer
 
 import logging
@@ -86,11 +91,16 @@ def _query_from_facets(self):
                         }
                     )
                 elif facet["type"] == "daterangeFacet":
+                    daterange = self.request.form[facet["field"]["value"]].split(",")
+                    if not daterange[0]:
+                        daterange[0] = "1970-01-01"
+                    if not daterange[1]:
+                        daterange[1] = "2500-01-01"
                     query.append(
                         {
                             "i": facet["field"]["value"],
                             "o": "plone.app.querystring.operation.date.between",
-                            "v": self.request.form[facet["field"]["value"]].split(","),
+                            "v": daterange,
                         }
                     )
                 elif facet["type"] == "selectFacet" and not facet["multiple"]:
@@ -110,7 +120,7 @@ def _query_from_facets(self):
                         }
                     )
                 else:
-                    logger.warning("DEBUG: filter %s not implemnted", facet)
+                    logger.warning("DEBUG: filter %s not implemented", facet)
                     query.append(
                         {
                             "i": facet["field"]["value"],
@@ -128,7 +138,6 @@ def get_data(self):
         # 4. fare la ricerca
         # 5. fare export in csv/pdf a seconda del formato
         """
-
         # 2. Get columns, base filters and sorting
         columns = self.block_data.get("columns", [])
 
@@ -185,13 +194,14 @@ def get_data(self):
         # XXX: consideriamo però che senza usare il serializzatore un utente potrebbe
         #      chiedere qualsiasi atttributo degli oggetti, senza un controllo fine
         #      sullo schema
-        fullobjects = True
-        self.request.form["b_size"] = 9999
-        results = getMultiAdapter((results, self.request), ISerializeToJson)(
-            fullobjects=fullobjects
-        )
-        for obj in results["items"]:
-            yield [obj["title"]] + [obj.get(c["field"]) for c in columns]
+        if results:
+            fullobjects = True
+            self.request.form["b_size"] = 9999
+            results = getMultiAdapter((results, self.request), ISerializeToJson)(
+                fullobjects=fullobjects
+            )
+            for obj in results["items"]:
+                yield [obj["title"]] + [obj.get(c["field"]) for c in columns]
 
     def get_columns(self, data):
         # Il titolo va aggiunto di default come prima colonna ?
@@ -200,3 +210,52 @@ def get_columns(self, data):
         return [{"key": "title", "title": _("Titolo")}] + [
             {"key": c["field"], "title": c["title"]} for c in columns
         ]
+
+    @view.memoize
+    def _get_querystring(self):
+        # @querystring endpoint
+        context = self.context.context
+        registry = getUtility(IRegistry)
+        reader = getMultiAdapter((registry, self.request), IQuerystringRegistryReader)
+        reader.vocab_context = context
+        result = reader()
+        return result
+
+    # TODO: valutare eventuale titolo impostato sul blocco
+    # def pdf_title(self):
+
+    pdf_description_as_html = True
+
+    def pdf_description(self) -> str:
+        query = []
+        querystring_registry = self._get_querystring()
+        searchtext = self._query_from_searchtext()
+        if searchtext and searchtext[0].get("v"):
+            # TODO: translate
+            query.append(f"Ricerca per: {searchtext[0]['v']}")
+        for facet in self.block_data.get("facets") or []:
+            if "field" not in facet:
+                logger.warning("invalid facet %s", facet)
+                continue
+            if facet["field"]["value"] in self.request.form:
+                value = self.request.form[facet["field"]["value"]]
+                if value in ["null"]:
+                    continue
+                # TODO: gestire campi particolari come: multipli, date, ...
+                index = querystring_registry["indexes"].get(facet["field"]["value"])
+                if index:
+                    if "values" in index:
+                        # TODO: per i valori multipli ?
+                        # TODO: facciamo constraint o fallback come ora?
+                        if value in index["values"] and index["values"][value].get(
+                            "title"
+                        ):
+                            query.append(
+                                f'{facet["field"]["label"]}: {index["values"][value]["title"]}'
+                            )
+                            continue
+                query.append(f'{facet["field"]["label"]}: {value}')
+        if query:
+            # TODO: translate
+            txt = "Filtri applicati:\n- " + ",\n- ".join(query)
+            return convertWebIntelligentPlainTextToHtml(txt)