Work around pdfium bug

VikParuchuri · May 6, 2024 · ebace90 · mara004 · May 7, 2024 · mara004
1 parent 54a4218
commit ebace90
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/pdftext/pdf/chars.py b/pdftext/pdf/chars.py
@@ -75,7 +75,7 @@ def get_pdfium_chars(pdf, fontname_sample_freq=settings.FONTNAME_SAMPLE_FREQ, pa
 
             rotation = pdfium_c.FPDFText_GetCharAngle(text_page, i)
             rotation = rotation * 180 / math.pi # convert from radians to degrees
-            coords = text_page.get_charbox(i, loose=True)
+            coords = text_page.get_charbox(i, loose=rotation == 0) # Loose doesn't work properly when page is rotated
             device_coords = page_bbox_to_device_bbox(page, coords, page_width, page_height, bl_origin, page_rotation, normalize=True)
 
             char_info = {

diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pdftext"
-version = "0.3.5"
+version = "0.3.6"
 description = "Extract structured text from pdfs quickly"
 authors = ["Vik Paruchuri <vik.paruchuri@gmail.com>"]
 license = "Apache-2.0"