v1.6.0 : var inside header or footer

.gitignore

@@ -18,6 +18,7 @@ exports/
 /lotemplate/unittest/files/content/debug.odt
 /lotemplate/unittest/files/content/debug.json
 /lotemplate/unittest/files/content/*.unittest.odt
+/lotemplate/unittest/files/content/*.unittest.pdf
 /venv
 /output*.*
 .fontconfig/

README.md

@@ -839,7 +839,10 @@ For trying to fix these problems, you can try:
 
 <a name="versions"></a>Versions
 -------------------------------
-
+- v1.6.0 : 2024-04-11
+  - allow put variables inside headers and footers
+  - fix a bug when a variable is both inside the text content and inside a table (it should not arrive, but it is fixed)
+  - a new unit test system based on PDF converted to text in order to test contents that are not converted to text with a simple saveAs 
 - v1.5.2 : 2024-02-24 : Better README
   - Rewrite for a betterdocker DockerFile without bug 
 - v1.5.1 : 2024-02-16 : Better README

lotemplate/Statement/TextStatement.py

@@ -7,7 +7,8 @@
 
 
 class TextStatement:
-    text_regex = regex.compile(r'\$(\w+(\(((?:\\.|.)*?)\))?)')
+    text_regex_as_string = r'\$(\w+(\(((?:\\.|.)*?)\))?)'
+    text_regex = regex.compile(text_regex_as_string)
     def __init__(self, text_string):
         self.text_string = text_string
 
@@ -19,11 +20,21 @@ def scan_text(doc: XComponent) -> dict[str, dict[str, str]]:
         :return: the scanned variables
         """
 
-        raw_string = doc.getText().getString()
-        matches = TextStatement.text_regex.finditer(raw_string)
+        search = doc.createSearchDescriptor()
+        search.SearchString = TextStatement.text_regex_as_string
+        search.SearchRegularExpression = True
+        search.SearchCaseSensitive = False
+        founded = doc.findAll(search)
+
+        simple_var_list = []
+        for x_found in founded:
+            text = x_found.getText()
+            cursor = text.createTextCursorByRange(x_found)
+            simple_var_list.append(cursor.String)
+
         plain_vars = {}
-        for var in matches:
-            key_name = var[0][1:]
+        for var in simple_var_list:
+            key_name = var[1:]
             # add to plain_vars if it doesn't matche ForStatement.foritem_regex
             if not re.search(ForStatement.forindex_regex, key_name, re.IGNORECASE):
                 plain_vars[key_name] = {'type': 'text', 'value': ''}
@@ -38,8 +49,11 @@ def scan_text(doc: XComponent) -> dict[str, dict[str, str]]:
                 text_fields_vars = (text_fields_vars |
                                     {var.group(0)[1:]: {'type': 'text', 'value': ''} for var in matches})
 
-        for var in TableStatement.scan_table(doc, get_list=True):
-            if '$' + var in plain_vars:
+        table_var_list = TableStatement.scan_table(doc, get_list=True)
+        for var in table_var_list:
+            if var.startswith("$"):
+                var = var[1:]
+            if var in plain_vars:
                 del plain_vars[var]
 
         for var in ForStatement.scan_for(doc):

lotemplate/unittest/files/content/text_var_in_header.expected.txt

@@ -0,0 +1,3 @@
+Can we change the following var my replaced text in the header ?
+This document test the replacement of a variable in the header.
+And in the footer the var “my replaced text” is still ther

lotemplate/unittest/files/content/text_var_in_header.json

@@ -0,0 +1,6 @@
+{
+  "my_var": {
+    "type": "text",
+    "value": "my replaced text"
+  }
+}

lotemplate/unittest/test_content.py

@@ -10,6 +10,7 @@
 import lotemplate as ot
 from time import sleep
 import subprocess
+from pypdf import PdfReader
 
 subprocess.call(f'soffice "--accept=socket,host=localhost,port=2002;urp;StarOffice.ServiceManager" &', shell=True)
 sleep(2)
@@ -28,7 +29,10 @@ def to_data(file: str):
     return ot.convert_to_datas_template(file_to_dict(file))
 
 
-def compare_files(name: str):
+def compare_files(name: str, format: str = 'txt'):
+    if format not in ['txt', 'pdf']:
+        return False
+
     base_path = 'lotemplate/unittest/files/content'
 
     def get_filename(ext: str):
@@ -50,17 +54,29 @@ def get_filename(ext: str):
     temp.search_error(to_data(get_filename('json')))
     temp.fill(file_to_dict(get_filename('json')))
 
-    if os.path.isfile(get_filename('unittest.txt')):
-        os.remove(get_filename('unittest.txt'))
-    temp.export(get_filename('unittest.txt'), True)
+    if os.path.isfile(get_filename('unittest.'+format)):
+        os.remove(get_filename('unittest.'+format))
+    temp.export(get_filename('unittest.'+format), True)
     # temp.close()
     if os.path.isfile(get_filename('unittest.odt')):
         os.remove(get_filename('unittest.odt'))
     temp.export(get_filename('unittest.odt'), True)
     temp.close()
+
+    # The PDF format is used to test some documents with headers or footers that are not supported by the text saveAs from
+    # LibreOffice. The PDF is then converted to text to compare with the expected text.
+    if format == 'pdf':
+        # convert to text
+        reader = PdfReader(get_filename('unittest.pdf'))
+        text = ""
+        for page in reader.pages:
+            text += page.extract_text() + "\n"
+        if os.path.isfile(get_filename('unittest.txt')):
+            os.remove(get_filename('unittest.txt'))
+        with open(get_filename('unittest.txt'), 'w') as f:
+            f.write(text)
+
     response = filecmp.cmp(get_filename('unittest.txt'), get_filename('expected.txt'))
-    # if os.path.isfile(get_filename('unittest.txt')):
-    #    os.remove(get_filename('unittest.txt'))
     return response
 
 
@@ -112,5 +128,8 @@ def test_image(self):
     def test_counter(self):
         self.assertTrue(compare_files('counter'))
 
+    def test_text_var_in_header(self):
+        self.assertTrue(compare_files('text_var_in_header', 'pdf'))
+
     def test_debug(self):
         self.assertTrue(compare_files('debug'))

lotemplate/unittest/test_template_scan.py

@@ -40,6 +40,13 @@ def test_format(self):
             (doc := ot.Template("lotemplate/unittest/files/templates/text_vars.odt", cnx, False)).scan())
         doc.close()
 
+    def test_text_var_in_header(self):
+        self.assertEqual(
+            {"my_var": {"type": "text", "value": ""}},
+            (doc := ot.Template("lotemplate/unittest/files/templates/text_var_in_header.odt", cnx, False)).scan()
+        )
+        doc.close()
+
     def test_static_table(self):
         self.assertEqual(
             {"var1": {"type": "text", "value": ""}, "var2": {"type": "text", "value": ""}},

requirements.txt

@@ -6,3 +6,4 @@ gunicorn~=20.1.0
 Werkzeug~=2.2.2
 sorcery~=0.2.1
 regex~=2022.10.31
+pypdf~=4.2.0