From 95f590db81dff2f7f10213f5ce63501a691deec5 Mon Sep 17 00:00:00 2001
From: monchin <pazzinter@126.com>
Date: Thu, 20 Nov 2025 09:55:10 +0800
Subject: [PATCH 1/2] Improve the performance of table-extraction by judging
 whether to do "make_chars" or "make_edges" by checking strategy

---
 src/table.py | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/src/table.py b/src/table.py
index 9b73782d3..2daad996f 100644
--- a/src/table.py
+++ b/src/table.py
@@ -2656,15 +2656,19 @@ def find_tables(
         tset = TableSettings.resolve(settings=settings)
         page.table_settings = tset
 
-        make_chars(page, clip=clip)  # create character list of page
-        make_edges(
-            page,
-            clip=clip,
-            tset=tset,
-            paths=paths,
-            add_lines=add_lines,
-            add_boxes=add_boxes,
-        )  # create lines and curves
+        if tset.vertical_strategy == "text" or tset.horizontal_strategy == "text":
+            make_chars(page, clip=clip)  # create character list of page
+        if tset.vertical_strategy.startswith(
+            "lines"
+        ) or tset.horizontal_strategy.startswith("lines"):
+            make_edges(
+                page,
+                clip=clip,
+                tset=tset,
+                paths=paths,
+                add_lines=add_lines,
+                add_boxes=add_boxes,
+            )  # create lines and curves
 
         tbf = TableFinder(page, settings=tset)
 

From 4aebd333838447e8f29d8ae599af55cb089a9308 Mon Sep 17 00:00:00 2001
From: monchin <pazzinter@126.com>
Date: Wed, 26 Nov 2025 16:34:03 +0800
Subject: [PATCH 2/2] Call make_chars before Table.extract if it is not called
 before

---
 src/table.py | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/src/table.py b/src/table.py
index 2daad996f..5becbc6b2 100644
--- a/src/table.py
+++ b/src/table.py
@@ -89,6 +89,7 @@
 
 EDGES = []  # vector graphics from PyMuPDF
 CHARS = []  # text characters from PyMuPDF
+CHARS_MADE = False # whether make_chars is executed. If not, make_chars is called before Table.extract()
 TEXTPAGE = None
 TEXT_BOLD = mupdf.FZ_STEXT_BOLD
 TEXT_STRIKEOUT = mupdf.FZ_STEXT_STRIKEOUT
@@ -1529,6 +1530,9 @@ def col_count(self) -> int:  # PyMuPDF extension
         return max([len(r.cells) for r in self.rows])
 
     def extract(self, **kwargs) -> list:
+        if not CHARS_MADE:
+            make_chars()
+
         chars = CHARS
         table_arr = []
 
@@ -2152,7 +2156,7 @@ def __getitem__(self, i):
 # -----------------------------------------------------------------------------
 def make_chars(page, clip=None):
     """Extract text as "rawdict" to fill CHARS."""
-    global TEXTPAGE
+    global TEXTPAGE, CHARS_MADE
     page_number = page.number + 1
     page_height = page.rect.height
     ctm = page.transformation_matrix
@@ -2204,6 +2208,7 @@ def make_chars(page, clip=None):
                         "y1": bbox_ctm.y1,
                     }
                     CHARS.append(char_dict)
+    CHARS_MADE = True
 
 
 # ------------------------------------------------------------------------
@@ -2586,9 +2591,10 @@ def find_tables(
     paths=None,  # accept vector graphics as parameter
 ):
     pymupdf._warn_layout_once()
-    global CHARS, EDGES
+    global CHARS, EDGES, CHARS_MADE
     CHARS = []
     EDGES = []
+    CHARS_MADE = False
     old_small = bool(pymupdf.TOOLS.set_small_glyph_heights())  # save old value
     pymupdf.TOOLS.set_small_glyph_heights(True)  # we need minimum bboxes
     if page.rotation != 0: