From f42fa4eba3c343cd3b9bcdfdf93bcd98353eec9a Mon Sep 17 00:00:00 2001
From: rogmann <github@rogmann.org>
Date: Wed, 28 Sep 2022 00:12:57 +0200
Subject: [PATCH 1/3] BUG: Operator Td involves a matrix multiplication.

---
 PyPDF2/_page.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/PyPDF2/_page.py b/PyPDF2/_page.py
index 459dae05c..66848bdb7 100644
--- a/PyPDF2/_page.py
+++ b/PyPDF2/_page.py
@@ -1490,8 +1490,13 @@ def process_operation(operator: bytes, operands: List) -> None:
             # Table 5.5 page 406
             elif operator == b"Td":
                 check_crlf_space = True
-                tm_matrix[4] += float(operands[0])
-                tm_matrix[5] += float(operands[1])
+                # A special case is a translating only tm:
+                # tm[0..5] = 1 0 0 1 e f,
+                # i.e. tm[4] += tx, tm[5] += ty.
+                tx = float(operands[0])
+                ty = float(operands[1])
+                tm_matrix[4] += tx * tm_matrix[0] + ty * tm_matrix[2]
+                tm_matrix[5] += tx * tm_matrix[1] + ty * tm_matrix[3]
             elif operator == b"Tm":
                 check_crlf_space = True
                 tm_matrix = [

From 84a6f5748a5967487e3af9db6a721ea8d633636b Mon Sep 17 00:00:00 2001
From: rogmann <github@rogmann.org>
Date: Wed, 28 Sep 2022 00:16:07 +0200
Subject: [PATCH 2/3] TST: Added test of Td-operation, adjusted test iss_1142.

---
 resources/Sample_Td-matrix.pdf | Bin 0 -> 1304 bytes
 tests/test_page.py             |  26 +++++++++++++++++++++++++-
 2 files changed, 25 insertions(+), 1 deletion(-)
 create mode 100644 resources/Sample_Td-matrix.pdf

diff --git a/resources/Sample_Td-matrix.pdf b/resources/Sample_Td-matrix.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..b3ea5de26930bd815fe88f75eaf34f929efc7b98
GIT binary patch
literal 1304
zcmaJ>QE%EX5PtWsxILh@njm%{g-sO?v}K#NY88$4vhBeT1F1q>Ifl~y`h90e$!OI?
z4(4+HzPs=4(z?AGw^gSnT5ViHkt9347Z(>4-Rgy@s0+(P5K)vqJs65EwbzARQ1r{V
zDz_y?I8DQ$FlELk;|9W|Gup%FN_#`6SD#}U$8r#NardPBDB}|;PtL&PY_^^mN2i;7
zb=Cl5@;aPLyDmK`ioWMrMZc-XfzjVOu3fZ;IQaG9Xz5Mq`BYM$x-nU<M|KOF#69T6
zlzd7U?^ZiFHxi$i%C6lEVz4Y@i<e=e=schehX@&QxP0{nU_(2Ov5_6eQiu4%lncKk
zne-F(s63=w3E7eKB|Leju-t-Ti7i(B>kN(H1L3juK9Kh<ABcitHE<Z?2nRWJ&TAjW
zdF2B#6b8ZYatkp^ImZ^=*o`4?ww}DDSRn$eO|B`;+V;V?&<jD1u_6^yI_E#NW18{5
z)32r|ETJp^$euc*^j_@1)71fk9&^96Yfsfo7l!5*M;=W^(<e}|JGU|DL+Fyy)q6a(
zQ}*FKHd<PjKcn=z*s%u4Jq?rXNR&)MLdVGvD)T;mwr(OvP$=#sdx;<;-~I615%m;x
zL3*SaYEJz<&vnWpxo@QXaB{!a9k&~2?q*IUx?4|u7;v_$U>fPlgl>;el#TKEOt(k2
z$i!jFw$982IE8$<=sZcf3C-!nrqJkoQF*WpP@*H7N(QfNu|m>Ee%%4Y_}_+cKx(zk
vJDnH!n(&a`<<ADfMc>LfdJ^eRD4JNy9}?b&QPa+cJd_;tMXPoFb1cL^id;cf

literal 0
HcmV?d00001

diff --git a/tests/test_page.py b/tests/test_page.py
index 5c78b14bb..6dbee6833 100644
--- a/tests/test_page.py
+++ b/tests/test_page.py
@@ -297,7 +297,14 @@ def test_iss_1142():
     name = "st2019.pdf"
     reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
     txt = reader.pages[3].extract_text()
-    assert txt.find("有限公司郑州分公司") > 0
+    # The following text is contained in two different cells:
+    # assert txt.find("有限公司郑州分公司") > 0
+    # 有限公司 = limited company
+    # 郑州分公司 = branch office in Zhengzhou
+    # First cell (see page 4/254):
+    assert txt.find("郑州药素电子商务有限公司") > 0
+    # Next cell (first cell in next line):
+    assert txt.find("郑州分公司") > 0
 
 
 @pytest.mark.parametrize(
@@ -604,6 +611,23 @@ def filter_first_table(r):
     assert texts.font_dict["/Encoding"] == "/WinAnsiEncoding"
     assert text_dat_of_date.font_size == 9.96
 
+    # Test 3: Read a table in a document using a non-translating
+    #         but scaling Tm-operand
+    reader = PdfReader(RESOURCE_ROOT / "Sample_Td-matrix.pdf")
+    page_td_model = reader.pages[0]
+    # We store the translations of the Td-executions.
+    list_Td = []
+    def visitor_td(op, args, cm, tm):
+        if op == b"Td":
+            list_Td.append((tm[4], tm[5]))
+    page_td_model.extract_text(visitor_operand_after=visitor_td)
+    assert len(list_Td) == 4
+    # Check the translations of the four Td-executions.
+    assert list_Td[0] == (210.0, 110.0)
+    assert list_Td[1] == (410.0, 110.0)
+    assert list_Td[2] == (210.0, 210.0)
+    assert list_Td[3] == (410.0, 210.0)
+
 
 @pytest.mark.parametrize(
     ("pdf_path", "password", "embedded", "unembedded"),

From e5be25b9016f7b398e8a678dffcaa9f43f49116e Mon Sep 17 00:00:00 2001
From: rogmann <github@rogmann.org>
Date: Wed, 28 Sep 2022 13:01:39 +0200
Subject: [PATCH 3/3] STY: Executed black.

---
 tests/test_page.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/test_page.py b/tests/test_page.py
index 6dbee6833..8516b666c 100644
--- a/tests/test_page.py
+++ b/tests/test_page.py
@@ -617,9 +617,11 @@ def filter_first_table(r):
     page_td_model = reader.pages[0]
     # We store the translations of the Td-executions.
     list_Td = []
+
     def visitor_td(op, args, cm, tm):
         if op == b"Td":
             list_Td.append((tm[4], tm[5]))
+
     page_td_model.extract_text(visitor_operand_after=visitor_td)
     assert len(list_Td) == 4
     # Check the translations of the four Td-executions.