pdftext 0.3.4__tar.gz → 0.3.6__tar.gz

{pdftext-0.3.4 → pdftext-0.3.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pdftext
-Version: 0.3.4
+Version: 0.3.6
 Summary: Extract structured text from pdfs quickly
 Home-page: https://github.com/VikParuchuri/pdftext
 License: Apache-2.0

{pdftext-0.3.4 → pdftext-0.3.6}/pdftext/pdf/chars.py

@@ -9,15 +9,13 @@ from pdftext.pdf.utils import get_fontname, pdfium_page_bbox_to_device_bbox, pag
 from pdftext.settings import settings
 
 
-def update_previous_fonts(text_chars: Dict, i: int, fontname: str, fontflags: int, prev_fontname: str, text_page, fontname_sample_freq: int):
-    min_update = max(0, i - fontname_sample_freq + 1) # Minimum index to update
-    regather_font_info = fontname != prev_fontname
-    for j in range(min_update, i): # Goes from min_update to i - 1
-        if regather_font_info:
-            fontname, fontflags = get_fontname(text_page, j)
+def update_previous_fonts(text_chars: Dict, i: int, prev_fontname: str, prev_fontflags: int, text_page, fontname_sample_freq: int):
+    min_update = max(0, i - fontname_sample_freq) # Minimum index to update
+    for j in range(i-1, min_update, -1): # Goes from i to min_update
+        fontname, fontflags = get_fontname(text_page, j)
 
         # If we hit the region with the previous fontname, we can bail out
-        if fontname == prev_fontname:
+        if fontname == prev_fontname and fontflags == prev_fontflags:
             break
         text_chars["chars"][j]["font"]["name"] = fontname
         text_chars["chars"][j]["font"]["flags"] = fontflags
@@ -70,12 +68,14 @@ def get_pdfium_chars(pdf, fontname_sample_freq=settings.FONTNAME_SAMPLE_FREQ, pa
             fontweight = round(pdfium_c.FPDFText_GetFontWeight(text_page, i), 1)
             if fontname is None or i % fontname_sample_freq == 0:
                 prev_fontname = fontname
+                prev_fontflags = fontflags
                 fontname, fontflags = get_fontname(text_page, i)
-                update_previous_fonts(text_chars, i, fontname, fontflags, prev_fontname, text_page, fontname_sample_freq)
+                if (fontname != prev_fontname or fontflags != prev_fontflags) and i > 0:
+                    update_previous_fonts(text_chars, i, prev_fontname, prev_fontflags, text_page, fontname_sample_freq)
 
             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 = {

{pdftext-0.3.4 → pdftext-0.3.6}/pdftext/settings.py

@@ -8,7 +8,7 @@ class Settings(BaseSettings):
     MODEL_PATH: str = os.path.join(BASE_PATH, "models", "dt.joblib")
 
     # Fonts
-    FONTNAME_SAMPLE_FREQ: int = 10
+    FONTNAME_SAMPLE_FREQ: int = 4
 
     # Inference
     BLOCK_THRESHOLD: float = 0.8 # Confidence threshold for block detection

{pdftext-0.3.4 → pdftext-0.3.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pdftext"
-version = "0.3.4"
+version = "0.3.6"
 description = "Extract structured text from pdfs quickly"
 authors = ["Vik Paruchuri <vik.paruchuri@gmail.com>"]
 license = "Apache-2.0"