From 5b43e1b50194b3f7dd228a0a8f60a835852fbd25 Mon Sep 17 00:00:00 2001
From: Josh Friedlander <16547083+lordgrenville@users.noreply.github.com>
Date: Wed, 3 Nov 2021 15:42:09 +0200
Subject: [PATCH 1/3] Detect CPORT header in SAS files

---
 pandas/io/sas/sas_xport.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/pandas/io/sas/sas_xport.py b/pandas/io/sas/sas_xport.py
index bdb7d86a9b37e..8f8c02a6c16ab 100644
--- a/pandas/io/sas/sas_xport.py
+++ b/pandas/io/sas/sas_xport.py
@@ -279,6 +279,10 @@ def _read_header(self):
         # read file header
         line1 = self._get_row()
         if line1 != _correct_line1:
+            if "**COMPRESSED**" in line1:
+                raise ValueError(
+                    "Header record indicates a CPORT file, which is not readable."
+                )
             raise ValueError("Header record is not an XPORT file.")
 
         line2 = self._get_row()

From 8485358c3cd686be7a1c86f39699329c75cbf687 Mon Sep 17 00:00:00 2001
From: Josh Friedlander <16547083+lordgrenville@users.noreply.github.com>
Date: Thu, 4 Nov 2021 14:08:29 +0200
Subject: [PATCH 2/3] add unit test for SAS CPORT check

---
 pandas/tests/io/sas/data/DEMO_PUF.cpt | Bin 0 -> 694 bytes
 pandas/tests/io/sas/test_xport.py     |   9 +++++++++
 2 files changed, 9 insertions(+)
 create mode 100644 pandas/tests/io/sas/data/DEMO_PUF.cpt

diff --git a/pandas/tests/io/sas/data/DEMO_PUF.cpt b/pandas/tests/io/sas/data/DEMO_PUF.cpt
new file mode 100644
index 0000000000000000000000000000000000000000..d74b6a70d281240e842f2ed568887e60dfac35b6
GIT binary patch
literal 694
zcmbWxyH3L}6b9hB&=NrhM%X+d6bTfkX{wGPNohoGvU{OZUw{#b6_s9qp&QH~QJ&%-
zc!E9?2{u)bGGW;|`laLV99?&lDCNEozNhO{|9{kfP8eKJH;MZ^3F)<MO_-5#)TwYp
z=iI_WfVI~#*eKywn5Kj7r0awsUWab8*`$=a%wl#Ma~_(uZDGN%6x5*V#tCKyo)}*4
zZf+T-sAg}2jYyq%JM^rLsbyf!u*k}Dh{vp~aDvajZP|Zj8km{Ceak$qVkh!J$roxG
zh*qgrDg=#czG>i9@^DO`m0G5>%zS2MnFWjFR?caX69Yc<$72psmEK3mZcKVid93Ii
zDsNAEL!XNviL?E7q_-~dZ0Xs`v$x(o$dU3I(sR6M(wjuSC;AS^t~B3D%$77;%4`))
ds?39;dD%<F2&5*-_Y%^Qq@_r!NS-Qk{~O4*-5~$~

literal 0
HcmV?d00001

diff --git a/pandas/tests/io/sas/test_xport.py b/pandas/tests/io/sas/test_xport.py
index 5d3e3b8e23cdb..9232ea8a25e4d 100644
--- a/pandas/tests/io/sas/test_xport.py
+++ b/pandas/tests/io/sas/test_xport.py
@@ -30,6 +30,7 @@ def setup_method(self, datapath):
         self.file02 = os.path.join(self.dirpath, "SSHSV1_A.xpt")
         self.file03 = os.path.join(self.dirpath, "DRXFCD_G.xpt")
         self.file04 = os.path.join(self.dirpath, "paxraw_d_short.xpt")
+        self.file05 = os.path.join(self.dirpath, "DEMO_PUF.cpt")
 
         with td.file_leak_context():
             yield
@@ -157,3 +158,11 @@ def test_truncated_float_support(self):
 
         data = read_sas(self.file04, format="xport")
         tm.assert_frame_equal(data.astype("int64"), data_csv)
+
+    def test_cport_header_found_raises(self):
+        # Test with DEMO_PUF.cpt, the beginning of puf2019_1_fall.xpt
+        # from https://www.cms.gov/files/zip/puf2019.zip
+        # (despite the extension, it's a cpt file)
+        msg = "Header record indicates a CPORT file, which is not readable."
+        with pytest.raises(ValueError, match=msg):
+            read_sas(self.file05, format="xport")

From 05f379dd23db139ae4ef3ede279d14b8d6ba3b37 Mon Sep 17 00:00:00 2001
From: Josh Friedlander <16547083+lordgrenville@users.noreply.github.com>
Date: Fri, 5 Nov 2021 16:00:47 +0200
Subject: [PATCH 3/3] add reference to documentation

---
 pandas/io/sas/sas_xport.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/pandas/io/sas/sas_xport.py b/pandas/io/sas/sas_xport.py
index 8f8c02a6c16ab..3f9bf6662e99f 100644
--- a/pandas/io/sas/sas_xport.py
+++ b/pandas/io/sas/sas_xport.py
@@ -280,6 +280,8 @@ def _read_header(self):
         line1 = self._get_row()
         if line1 != _correct_line1:
             if "**COMPRESSED**" in line1:
+                # this was created with the PROC CPORT method and can't be read
+                # https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/movefile/p1bm6aqp3fw4uin1hucwh718f6kp.htm
                 raise ValueError(
                     "Header record indicates a CPORT file, which is not readable."
                 )