From 8b9fdc8e71ccdbafceadc28ebd1a5280a2a6bc4f Mon Sep 17 00:00:00 2001
From: rbenes <radek.benes@gmail.com>
Date: Mon, 4 Feb 2019 14:00:44 +0100
Subject: [PATCH] Backport PR #25058: Reading a HDF5 created in py2

---
 doc/source/whatsnew/v0.24.2.rst                |   2 +-
 pandas/io/pytables.py                          |   2 +-
 .../io/data/legacy_hdf/legacy_table_py2.h5     | Bin 0 -> 72279 bytes
 pandas/tests/io/test_pytables.py               |  17 ++++++++++++++++-
 4 files changed, 18 insertions(+), 3 deletions(-)
 create mode 100644 pandas/tests/io/data/legacy_hdf/legacy_table_py2.h5

diff --git a/doc/source/whatsnew/v0.24.2.rst b/doc/source/whatsnew/v0.24.2.rst
index cba21ce7ee1e6..316fc21c126ac 100644
--- a/doc/source/whatsnew/v0.24.2.rst
+++ b/doc/source/whatsnew/v0.24.2.rst
@@ -51,7 +51,7 @@ Bug Fixes
 
 **I/O**
 
--
+- Bug in reading a HDF5 table-format ``DataFrame`` created in Python 2, in Python 3 (:issue:`24925`)
 -
 -
 
diff --git a/pandas/io/pytables.py b/pandas/io/pytables.py
index 4e103482f48a2..2ab6ddb5b25c7 100644
--- a/pandas/io/pytables.py
+++ b/pandas/io/pytables.py
@@ -3288,7 +3288,7 @@ def get_attrs(self):
         self.nan_rep = getattr(self.attrs, 'nan_rep', None)
         self.encoding = _ensure_encoding(
             getattr(self.attrs, 'encoding', None))
-        self.errors = getattr(self.attrs, 'errors', 'strict')
+        self.errors = _ensure_decoded(getattr(self.attrs, 'errors', 'strict'))
         self.levels = getattr(
             self.attrs, 'levels', None) or []
         self.index_axes = [
diff --git a/pandas/tests/io/data/legacy_hdf/legacy_table_py2.h5 b/pandas/tests/io/data/legacy_hdf/legacy_table_py2.h5
new file mode 100644
index 0000000000000000000000000000000000000000..3863d714a315b79de100e8446db766ec7dc2e82c
GIT binary patch
literal 72279
zcmeI#%}yIv6u|K@5O5q*2$iaeZdlb`Axyqlh89yS8G|rwQmN8(j3>yd8C$l4(p{JB
zTXdNR=^~HPw`k9~cMcu~oF-KlRq`M3jOWh%IP;r(?_B?Ou(!WlS}z5<6^p?_u<E|r
z?yq!vwq9^ac5c^yOuzW6LGWvO{(8=TDFlmwUi&;vUrp=ZnD+^Prs=0A^?D#Vt$sVE
zmxDKp{#(%!IDS+D2ld@!z2WESU0W}GllC*im_(g^EGO0<5A4FfJl;?HYrbEXgQ}}h
z`*%_PP8+99>A=CK^3k@feV0jg`TSouciU}A*Zg<nWBI>1Ir>&_bedmlF9zS$Puh*6
zmdme0$MV-Ux8zLo@3><@Mw;e7{{D3Lt7g3&?jM~T?w*FT%5QJh(sFCA&ipuCO$XfS
zg3A{C(>UE(_W9Cc{)GKUBQR^64%7MDApYxd+HM`~6{H7SzbmHw)yBzIdVZXX!{XuS
z1#G^{drjw~|J!_5cJtjZO1jZFocui0+27Br_vC2L|HUY}h(qgjI$!#Evk<Ig%U{N$
zac_|L^4e9b{MKge(dKq`p1tSFxf-X<y7^h(O;#=%4r7T-UP_@0p&T5@RlV-Z_mXb>
zDm)wX$1d5G|H57O=hgnOR;o6(O6`ivS}6^;OOxS)QncwZl!K@4Sdcr?dH*u%zl_Jz
z3ibW6CLfE()z_{>l~UBMbY$<Vk<u2I_ecgw=wpjs#S)v=vyrc-+K|}#5>U26AKWyq
zWfxaBnnW7a%ZtR_>AS~Ol~y%={~k=DBpk)V8}C6$cloBz^{y)sCuf6hFFAJ?WvA`x
zlEFM$SNCu2EctSy(O@KzYW{3L7*9sMvq`r9EIIjpd>QLA)c2+Q%l6dmtsOh&<GY9_
zk=nS-vdg#Uo|nW}4u)xccU*kE<i$V}H2<M|to~|O`inl6u6ryx9;C}WkSEm8o+jTP
zmd5Q$BBNm}Cy(^`JubDPu_Rgtd!nUxN@Jzb^@P@4OdGKE`!AnJqHM3zb^LtryUX{P
zoZpvpSC`B2_FdM?5WL^)a(d4iPAAQBcXn2HCEJGTY1z4XcaSrGfAe#5RUId?t*rQ4
za3^c&)6_D5eA)iNCCV=UzS$_-6^`#Cp#4j?LP3T7*wmx_{>#62X04muYFEll^ZWT?
z_PY0pHY52r+|jS&7jCTMWv<ZaLh{-B!p~>>je2u0+zMN}hq~-{<1o99xBh>7x!SdI
zdguP`gCNq%WV^lG_RMlN7N)-gmZgqjU~%p@noX@YI~MfZ+STv6-QI33wDLCgZkDs(
zAK7-dugbl9yV>{RPr~Tq(sq~RW9{~mNqioUGV%I(I|_Xn&96h!@@L+AZpNSX6Dy}P
zv44*Gz^&ZP_k-_OQP_2BjeK1Cen`%DZp77jdt6s<;(T14@crWIWa7HsA199=Iuq;1
zixOioXq_BAYfJj9=i@rRT-V=yy1uq?IX%k7Tej!x*AsP8)lJ_GYzQEL00Iag@L>x?
zwt%n&RVV$y_+j5RcZUE12q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*sr
zAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_0
z00IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@
z2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000Iag
zfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}
z0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*srAb<b@2q1s}0tg_000IagfB*sr
XAb<b@2q1s}0tg_000Iag;068$pZG^R

literal 0
HcmV?d00001

diff --git a/pandas/tests/io/test_pytables.py b/pandas/tests/io/test_pytables.py
index 517a3e059469c..9430011288f27 100644
--- a/pandas/tests/io/test_pytables.py
+++ b/pandas/tests/io/test_pytables.py
@@ -4540,7 +4540,7 @@ def test_pytables_native2_read(self, datapath):
 
     def test_legacy_table_fixed_format_read_py2(self, datapath):
         # GH 24510
-        # legacy table with fixed format written en Python 2
+        # legacy table with fixed format written in Python 2
         with ensure_clean_store(
                 datapath('io', 'data', 'legacy_hdf',
                          'legacy_table_fixed_py2.h5'),
@@ -4552,6 +4552,21 @@ def test_legacy_table_fixed_format_read_py2(self, datapath):
                                                    name='INDEX_NAME'))
             assert_frame_equal(expected, result)
 
+    def test_legacy_table_read_py2(self, datapath):
+        # issue: 24925
+        # legacy table written in Python 2
+        with ensure_clean_store(
+                datapath('io', 'data', 'legacy_hdf',
+                         'legacy_table_py2.h5'),
+                mode='r') as store:
+            result = store.select('table')
+
+        expected = pd.DataFrame({
+            "a": ["a", "b"],
+            "b": [2, 3]
+        })
+        assert_frame_equal(expected, result)
+
     def test_legacy_table_read(self, datapath):
         # legacy table types
         with ensure_clean_store(