No ligature preservation in searches

Specifying ligatures in search needles makes no sense. Not setting the flag bit causes text to be found with and without ligatures.
pymupdf · Jul 15, 2024 · 2414c80 · 2414c80
1 parent 3a259dc
commit 2414c80
Show file tree

Hide file tree

Showing 5 changed files with 17 additions and 3 deletions.
diff --git a/docs/app1.rst b/docs/app1.rst
@@ -286,7 +286,7 @@ Text Extraction Flags Defaults
 ========================= ==== ==== ===== === ==== ======= ===== ====== ======
 Indicator                 text html xhtml xml dict rawdict words blocks search
 ========================= ==== ==== ===== === ==== ======= ===== ====== ======
-preserve ligatures        1    1    1     1   1    1       1     1       1
+preserve ligatures        1    1    1     1   1    1       1     1       0
 preserve whitespace       1    1    1     1   1    1       1     1       1
 preserve images           n/a  1    1     n/a 1    1       n/a   0       0
 inhibit spaces            0    0    0     0   0    0       0     0       0

diff --git a/docs/vars.rst b/docs/vars.rst
@@ -262,7 +262,7 @@ The following constants represent the default combinations of the above for text
 
 .. py:data:: TEXTFLAGS_SEARCH
 
-    `TEXT_PRESERVE_LIGATURES | TEXT_PRESERVE_WHITESPACE | TEXT_MEDIABOX_CLIP | TEXT_DEHYPHENATE`
+    `TEXT_PRESERVE_WHITESPACE | TEXT_MEDIABOX_CLIP | TEXT_DEHYPHENATE`
 
 
 .. _linkDest Kinds:

diff --git a/src/__init__.py b/src/__init__.py
@@ -13312,7 +13312,6 @@ def width(self):
 TEXTFLAGS_RAWDICT = TEXTFLAGS_DICT
 
 TEXTFLAGS_SEARCH = (0
-        | TEXT_PRESERVE_LIGATURES
         | TEXT_PRESERVE_WHITESPACE
         | TEXT_MEDIABOX_CLIP
         | TEXT_DEHYPHENATE

diff --git a/tests/resources/text-find-ligatures.pdf b/tests/resources/text-find-ligatures.pdf
diff --git a/tests/test_textsearch.py b/tests/test_textsearch.py
@@ -7,13 +7,15 @@
 Text search with 'clip' parameter - clip rectangle contains two occurrences
 of searched text. Confirm search locations are inside clip.
 """
+
 import os
 
 import pymupdf
 
 scriptdir = os.path.abspath(os.path.dirname(__file__))
 filename1 = os.path.join(scriptdir, "resources", "2.pdf")
 filename2 = os.path.join(scriptdir, "resources", "github_sample.pdf")
+filename3 = os.path.join(scriptdir, "resources", "text-find-ligatures.pdf")
 
 
 def test_search1():
@@ -35,3 +37,16 @@ def test_search2():
     assert len(rl) == 2
     for r in rl:
         assert r in clip
+
+
+def test_search3():
+    """Ensure we find text whether or not it contains ligatures."""
+    doc = pymupdf.open(filename3)
+    page = doc[0]
+    needle = "flag"
+    hits = page.search_for(needle, flags=pymupdf.TEXTFLAGS_SEARCH)
+    assert len(hits) == 2  # all occurrences found
+    hits = page.search_for(
+        needle, flags=pymupdf.TEXTFLAGS_SEARCH | pymupdf.TEXT_PRESERVE_LIGATURES
+    )
+    assert len(hits) == 1  # only found text without ligatures