From 7e2893106e75fefdfddc967a0dfadb1573840445 Mon Sep 17 00:00:00 2001
From: Jeremy Singer-Vine <jsvine@gmail.com>
Date: Wed, 19 Jul 2023 08:56:18 -0400
Subject: [PATCH] Add `antialias` parameter to `Page.to_image(...)`

... and associated methods. Thanks to @cmdlineluser for flagging.

For details, see:
https://github.com/jsvine/pdfplumber/discussions/899#discussioncomment-6464765
---
 README.md             | 7 ++++---
 pdfplumber/display.py | 7 ++++++-
 pdfplumber/page.py    | 5 ++++-
 tests/test_display.py | 4 ++++
 4 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/README.md b/README.md
index e12589eb..bf7b3d70 100644
--- a/README.md
+++ b/README.md
@@ -255,9 +255,10 @@ If you pass the `pdfminer.six`-handling `laparams` parameter to `pdfplumber.open
 
 To turn any page (including cropped pages) into an `PageImage` object, call `my_page.to_image()`. You can optionally pass *one* of the  following keyword arguments:
 
-- `resolution`: The desired number pixels per inch. Defaults to 72.
-- `width`: The desired image width in pixels.
-- `height`: The desired image width in pixels.
+- `resolution`: The desired number pixels per inch. Default: `72`. Type: `int`.
+- `width`: The desired image width in pixels. Default: unset, determined by `resolution`. Type: `int`.
+- `height`: The desired image width in pixels. Default: unset, determined by `resolution`. Type: `int`.
+- `antialias`: Whether to use antialiasing when creating the image. Setting to `True` creates images with less-jagged text and graphics, but with larger file sizes. Default: `False`. Type: `bool`.
 
 For instance:
 
diff --git a/pdfplumber/display.py b/pdfplumber/display.py
index b5cd9acf..9b826424 100644
--- a/pdfplumber/display.py
+++ b/pdfplumber/display.py
@@ -38,6 +38,7 @@ def get_page_image(
     page_ix: int,
     resolution: Union[int, float],
     password: Optional[str],
+    antialias: bool = False,
 ) -> PIL.Image.Image:
     # If we are working with a file object saved to disk
     if hasattr(stream, "name"):
@@ -54,7 +55,9 @@ def get_page_image(
         input_data=src,
         password=password,
         scale=resolution / 72,
-        no_smoothtext=True,
+        no_smoothtext=not antialias,
+        no_smoothpath=not antialias,
+        no_smoothimage=not antialias,
         # Non-modifiable arguments
         renderer=pypdfium2._helpers.page.PdfPage.render,
         converter=pypdfium2.PdfBitmap.to_pil,
@@ -73,6 +76,7 @@ def __init__(
         page: "Page",
         original: Optional[PIL.Image.Image] = None,
         resolution: Union[int, float] = DEFAULT_RESOLUTION,
+        antialias: bool = False,
     ):
         self.page = page
         if original is None:
@@ -80,6 +84,7 @@ def __init__(
                 stream=page.pdf.stream,
                 page_ix=page.page_number - 1,
                 resolution=resolution,
+                antialias=antialias,
                 password=page.pdf.password,
             )
         else:
diff --git a/pdfplumber/page.py b/pdfplumber/page.py
index 87df397c..521002f4 100644
--- a/pdfplumber/page.py
+++ b/pdfplumber/page.py
@@ -460,6 +460,7 @@ def to_image(
         resolution: Optional[Union[int, float]] = None,
         width: Optional[Union[int, float]] = None,
         height: Optional[Union[int, float]] = None,
+        antialias: bool = False,
     ) -> "PageImage":
         """
         You can pass a maximum of 1 of the following:
@@ -479,7 +480,9 @@ def to_image(
         elif height is not None:
             resolution = 72 * height / self.height
 
-        return PageImage(self, resolution=resolution or DEFAULT_RESOLUTION)
+        return PageImage(
+            self, resolution=resolution or DEFAULT_RESOLUTION, antialias=antialias
+        )
 
     def to_dict(self, object_types: Optional[List[str]] = None) -> Dict[str, Any]:
         if object_types is None:
diff --git a/tests/test_display.py b/tests/test_display.py
index dcb7a54c..88bc0cd5 100644
--- a/tests/test_display.py
+++ b/tests/test_display.py
@@ -97,6 +97,10 @@ def test_no_quantize(self):
         self.im.save(b, "PNG", quantize=False)
         assert len(b.getvalue()) > len(self.im._repr_png_())
 
+    def test_antialias(self):
+        aa = self.pdf.pages[0].to_image(antialias=True)
+        assert len(aa._repr_png_()) > len(self.im._repr_png_())
+
     def test_decompression_bomb(self):
         original_max = PIL.Image.MAX_IMAGE_PIXELS
         PIL.Image.MAX_IMAGE_PIXELS = 10