diff tests/test_rtl.py @ 1:1d09e1dec1d9 upstream

ADD: PyMuPDF v1.26.4: the original sdist. It does not yet contain MuPDF. This normally will be downloaded when building PyMuPDF.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:37:51 +0200
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/test_rtl.py	Mon Sep 15 11:37:51 2025 +0200
@@ -0,0 +1,18 @@
+import pymupdf
+
+import os
+
+
+def test_rtl():
+    path = os.path.normpath(f'{__file__}/../../tests/resources/test-E+A.pdf')
+    doc = pymupdf.open(path)
+    page = doc[0]
+    # set of all RTL characters
+    rtl_chars = set([chr(i) for i in range(0x590, 0x901)])
+
+    for w in page.get_text("words"):
+        # every word string must either ONLY contain RTL chars
+        cond1 = rtl_chars.issuperset(w[4])
+        # ... or NONE.
+        cond2 = rtl_chars.intersection(w[4]) == set()
+        assert cond1 or cond2