Merge pull request #371 from chinapandaman/PPF-370

PPF-370: implement smart line break

PyPDFForm/core/filler.py

@@ -48,7 +48,7 @@ def fill(
             else:
                 elements[
                     key
-                ].last_line_x_coordinate = template.get_last_line_x_coordinate(
+                ].text_line_x_coordinates = template.get_text_line_x_coordinates(
                     _element, elements[key]
                 )
                 x, y = template.get_draw_text_coordinates(_element, elements[key])

PyPDFForm/core/template.py

@@ -327,23 +327,27 @@ def get_draw_text_coordinates(
     return x, y
 
 
-def get_last_line_x_coordinate(
+def get_text_line_x_coordinates(
     element: pdfrw.PdfDict, element_middleware: Text
-) -> Union[float, int, None]:
+) -> Union[List[float], None]:
     """
-    Returns the x coordinate to draw the last line
+    Returns the x coordinates to draw lines
     of the text at given a PDF form paragraph element.
     """
 
     if (
         element_middleware.text_wrap_length is not None
+        and element_middleware.text_lines is not None
+        and len(element_middleware.text_lines)
         and isinstance(element_middleware.value, str)
         and len(element_middleware.value) > element_middleware.text_wrap_length
     ):
+        result = []
         _ele = deepcopy(element_middleware)
-        last_line_mark = -1 * (len(_ele.value) % _ele.text_wrap_length)
-        _ele.value = _ele.value[last_line_mark:]
+        for each in element_middleware.text_lines:
+            _ele.value = each
+            result.append(get_draw_text_coordinates(element, _ele)[0])
 
-        return get_draw_text_coordinates(element, _ele)[0]
+        return result
 
     return None

PyPDFForm/core/utils.py

@@ -2,7 +2,7 @@
 """Contains utility helpers."""
 
 from io import BytesIO
-from typing import Dict, Union
+from typing import Dict, Union, List
 
 import pdfrw
 from reportlab.pdfbase.pdfmetrics import stringWidth
@@ -47,10 +47,50 @@ def update_text_field_attributes(
                     elements[key].font_size = template.get_text_field_font_size(
                         _element
                     ) or font_size_core.text_field_font_size(_element)
-                if template.is_text_multiline(_element):
+                if template.is_text_multiline(_element) and elements[key].text_wrap_length is None:
                     elements[key].text_wrap_length = get_paragraph_auto_wrap_length(
                         _element, elements[key]
                     )
+                    elements[key].text_lines = get_paragraph_lines(elements[key])
+
+
+def get_paragraph_lines(
+    element_middleware: Text
+) -> List[str]:
+    """Splits the paragraph field's text to a list of lines."""
+
+    lines = []
+    result = []
+    text_wrap_length = element_middleware.text_wrap_length
+    value = element_middleware.value or ""
+    if element_middleware.max_length is not None:
+        value = value[:element_middleware.max_length]
+    characters = value.split(" ")
+    current_line = ""
+    for each in characters:
+        line_extended = f"{current_line} {each}" if current_line else each
+        if len(line_extended) <= text_wrap_length:
+            current_line = line_extended
+        else:
+            lines.append(current_line)
+            current_line = each
+    lines.append(current_line)
+
+    for each in lines:
+        while len(each) > text_wrap_length:
+            last_index = text_wrap_length - 1
+            result.append(each[:last_index])
+            each = each[last_index:]
+        if each:
+            if result and len(each) + 1 + len(result[-1]) <= text_wrap_length:
+                result[-1] = f"{result[-1]}{each} "
+            else:
+                result.append(f"{each} ")
+
+    if result:
+        result[-1] = result[-1][:-1]
+
+    return result
 
 
 def get_paragraph_auto_wrap_length(

PyPDFForm/core/watermark.py

@@ -58,22 +58,17 @@ def draw_text(
         )
     else:
         text_obj = canv.beginText(0, 0)
-
-        start = 0
-        end = element.text_wrap_length
-
-        while end < len(text_to_draw):
-            text_obj.textLine(text_to_draw[start:end])
-            start += element.text_wrap_length
-            end += element.text_wrap_length
-
-        if (
-            element.last_line_x_coordinate is not None
-            and element.last_line_x_coordinate - coordinate_x != 0
-        ):
-            text_obj.moveCursor(element.last_line_x_coordinate - coordinate_x, 0)
-
-        text_obj.textLine(text_to_draw[start:])
+        for i, line in enumerate(element.text_lines):
+            cursor_moved = False
+            if (
+                element.text_line_x_coordinates is not None
+                and element.text_line_x_coordinates[i] - coordinate_x != 0
+            ):
+                text_obj.moveCursor(element.text_line_x_coordinates[i] - coordinate_x, 0)
+                cursor_moved = True
+            text_obj.textLine(line)
+            if cursor_moved:
+                text_obj.moveCursor(-1 * (element.text_line_x_coordinates[i] - coordinate_x), 0)
 
         canv.saveState()
         canv.translate(

PyPDFForm/middleware/text.py

@@ -25,7 +25,8 @@ def __init__(
         self.max_length = None
         self.comb = None
         self.character_paddings = None
-        self.last_line_x_coordinate = None
+        self.text_lines = None
+        self.text_line_x_coordinates = None
 
     @property
     def schema_definition(self) -> dict:

PyPDFForm/wrapper.py

@@ -43,7 +43,6 @@ def __init__(
                 each.text_y_offset = kwargs.get(
                     "global_text_y_offset", constants.GLOBAL_TEXT_Y_OFFSET
                 )
-                each.text_wrap_length = kwargs.get("global_text_wrap_length")
 
     def read(self) -> bytes:
         """Reads the file stream of a PDF form."""
@@ -109,7 +108,6 @@ def draw_text(
         new_element.text_y_offset = kwargs.get(
             "text_y_offset", constants.GLOBAL_TEXT_Y_OFFSET
         )
-        new_element.text_wrap_length = kwargs.get("text_wrap_length")
 
         watermarks = watermark_core.create_watermarks_and_draw(
             self.stream,

docs/examples.md

@@ -131,41 +131,6 @@ with open(PATH_TO_FILLED_PDF_FORM, "wb+") as output:
     )
 ```
 
-## Wrap filled text with a length
-
-Sometimes texts printed on the PDF form may be too lengthy. This example 
-demos globally wrapping texts with a given length.
-
-```python
-import os
-
-from PyPDFForm import PyPDFForm
-
-PATH_TO_DOWNLOADED_SAMPLE_PDF_FORM = os.path.join(
-    os.path.expanduser("~/Downloads"), "sample_template.pdf"
-)  # Change this to where you downloaded the sample PDF form
-
-PATH_TO_FILLED_PDF_FORM = os.path.join(
-    os.path.expanduser("~"), "output.pdf"
-)  # Change this to where you wish to put your filled PDF form
-
-with open(PATH_TO_FILLED_PDF_FORM, "wb+") as output:
-    output.write(
-        PyPDFForm(PATH_TO_DOWNLOADED_SAMPLE_PDF_FORM, global_text_wrap_length=2,)
-        .fill(
-            {
-                "test": "test_1",
-                "check": True,
-                "test_2": "test_2",
-                "check_2": False,
-                "test_3": "test_3",
-                "check_3": True,
-            },
-        )
-        .read()
-    )
-```
-
 ## Offset texts globally
 
 This example offsets all texts printed on the PDF form by 100 horizontally 
@@ -237,10 +202,8 @@ with open(PATH_TO_FILLED_PDF_FORM, "wb+") as output:
     pdf_form.elements["test"].font_color = (1, 0, 0)
     pdf_form.elements["test_2"].text_x_offset = 50
     pdf_form.elements["test_2"].text_y_offset = -50
-    pdf_form.elements["test_2"].text_wrap_length = 1
     pdf_form.elements["test_2"].font_color = (0, 1, 0)
     pdf_form.elements["test_3"].font = "LiberationSerif-Italic"
-    pdf_form.elements["test_3"].text_wrap_length = 2
     pdf_form.elements["test_3"].font_color = (0, 0, 1)
 
     pdf_form.fill(

tests/conftest.py

@@ -86,6 +86,14 @@ def sample_template_paragraph_complex(pdf_samples):
         return f.read()
 
 
+@pytest.fixture
+def sample_template_with_paragraph_max_length(pdf_samples):
+    with open(
+        os.path.join(pdf_samples, "sample_template_with_paragraph_max_length.pdf"), "rb+"
+    ) as f:
+        return f.read()
+
+
 @pytest.fixture
 def sample_template_with_max_length_text_field(pdf_samples):
     with open(