From 9fcd30bda773064c089815245521a1237e89ce22 Mon Sep 17 00:00:00 2001
From: Alex Rothberg <agrothberg@gmail.com>
Date: Fri, 16 Aug 2013 19:47:14 -0400
Subject: [PATCH] ENH Factored out excel_value_to_python_value from
 ExcelFile::_parse_excel (GH4589)

---
 doc/source/release.rst         |   3 ++-
 pandas/io/excel.py             |  31 ++++++++++++++++++-------------
 pandas/io/tests/data/types.xls | Bin 0 -> 16384 bytes
 pandas/io/tests/test_excel.py  |  29 ++++++++++++++++++++++++++++-
 4 files changed, 48 insertions(+), 15 deletions(-)
 create mode 100755 pandas/io/tests/data/types.xls

diff --git a/doc/source/release.rst b/doc/source/release.rst
index a2b525a737879..275717ef984ea 100644
--- a/doc/source/release.rst
+++ b/doc/source/release.rst
@@ -44,7 +44,8 @@ pandas 0.13
   - Text parser now treats anything that reads like inf ("inf", "Inf", "-Inf",
     "iNf", etc.) to infinity. (:issue:`4220`, :issue:`4219`), affecting
     ``read_table``, ``read_csv``, etc.
-
+  - Factored out excel_value_to_python_value from ExcelFile::_parse_excel (:issue:`4589`)
+  
 **API Changes**
 
   - ``pandas`` now is Python 2/3 compatible without the need for 2to3 thanks to
diff --git a/pandas/io/excel.py b/pandas/io/excel.py
index 534a88e303dbf..aa89264cfc695 100644
--- a/pandas/io/excel.py
+++ b/pandas/io/excel.py
@@ -53,6 +53,23 @@ def read_excel(path_or_buf, sheetname, kind=None, **kwds):
     return ExcelFile(path_or_buf, kind=kind).parse(sheetname=sheetname,
                                                    kind=kind, **kwds)
 
+def excel_value_to_python_value(value, typ, datemode):
+    from xlrd import (xldate_as_tuple, XL_CELL_DATE,
+                      XL_CELL_ERROR, XL_CELL_BOOLEAN)
+            
+    if typ == XL_CELL_DATE:
+        dt = xldate_as_tuple(value, datemode)
+        # how to produce this first case?
+        if dt[0] < datetime.MINYEAR:  # pragma: no cover
+            value = datetime.time(*dt[3:])
+        else:
+            value = datetime.datetime(*dt)
+    elif typ == XL_CELL_ERROR:
+        value = np.nan
+    elif typ == XL_CELL_BOOLEAN:
+        value = bool(value)
+    
+    return value
 
 class ExcelFile(object):
     """
@@ -174,8 +191,6 @@ def _parse_excel(self, sheetname, header=0, skiprows=None, skip_footer=0,
                      index_col=None, has_index_names=None, parse_cols=None,
                      parse_dates=False, date_parser=None, na_values=None,
                      thousands=None, chunksize=None, **kwds):
-        from xlrd import (xldate_as_tuple, XL_CELL_DATE,
-                          XL_CELL_ERROR, XL_CELL_BOOLEAN)
 
         datemode = self.book.datemode
         if isinstance(sheetname, compat.string_types):
@@ -193,17 +208,7 @@ def _parse_excel(self, sheetname, header=0, skiprows=None, skip_footer=0,
                     should_parse[j] = self._should_parse(j, parse_cols)
 
                 if parse_cols is None or should_parse[j]:
-                    if typ == XL_CELL_DATE:
-                        dt = xldate_as_tuple(value, datemode)
-                        # how to produce this first case?
-                        if dt[0] < datetime.MINYEAR:  # pragma: no cover
-                            value = datetime.time(*dt[3:])
-                        else:
-                            value = datetime.datetime(*dt)
-                    elif typ == XL_CELL_ERROR:
-                        value = np.nan
-                    elif typ == XL_CELL_BOOLEAN:
-                        value = bool(value)
+                    value = excel_value_to_python_value(value=value, typ=typ, datemode=datemode)
                     row.append(value)
 
             data.append(row)
diff --git a/pandas/io/tests/data/types.xls b/pandas/io/tests/data/types.xls
new file mode 100755
index 0000000000000000000000000000000000000000..232a88609df92754e4bb8759825ff0361874d73b
GIT binary patch
literal 16384
zcmeHOdvH|M89#S-lWYi*1VTU_VF|A!BoGpw@>m`|$M6`%!HzhX$1bGAgsvM4Gg3lp
z|Ebi$q(TK+Yn7Q+tyZ)?rnW-I(ROCFomOkH+QAMwwmNO0c3K@_`}>`<clX|NADbO#
zT6H+#+`Z?!=ls6!eCIpgdGecISDw1>u?cSpEvyhhaXwutiY;^l?~F232r&^aOrKAu
z(~2krUN3~tkOs~{*2!qRAYury2$5}Gf*3|DMI3=xhBy-OBE(UM<%lo=QHeMjaSUP=
z;#kCL#Bqq@5idrZfJhr*tg*a*hSc=A74;g?hrgtVh|5GTeiLFBT4?y88Wv+JhrS>Z
z4#>jC6~yK0$NrqeB#jMW6SybDcF`f_*sh2LJV%Rwgd$hUF`S~zyJ()uStPC%{o-c)
zwadC+1uev-Or#`5H{N@FK<mJteA$17%b#$+%wQIYOTnR2$`F$+P?$W6GyKV$Ap`6C
zL)Cwf$x#|cFD@<;4^DZh<OIBVF+RQp%pZqN;3>qP5jTq+e2@Gu!)hr*yFnR_!?2d)
z%@9XGt$<G<l`T)d!{?(i-?Q?$9A8cuO`>AklIF$u#xd&pz)G3#oAOyJpQG?J#8a}&
z>yk20)QC5bKP>jslmnun#8?9>=z$e<z&g6crzyua!w`$bv#fqr(wj4Kwo6GHU2@p#
z63Uz74d_VcV#lDyD)VZrDsPTud2_7Dlfw{EO&_|XhM1oxhd>Vh);BOF+O@88;fU#v
z5y1H9pTmF^=HpW8iBS`p1iy?ixORH&w8m-Enzz(1-CSoTHrLmLrec(bA2vz&e#14f
zt{d%{YB3GK<iN}{r`e8!G@)ie)LK24hL*{?NHmBRA+}JIhKMOyCb_vz5;oT_%@Asv
zY9ZH_x|LhNWsa!FABtMbHAJk18X~q5QPC(au`^ROu=A6mSy)D+Ii?H6Z1kiYQ?q=Q
zuUS~K&XqhR=R!jKD;+`aFvIZ(>2xiUZn|LMPiF^r%kM=eH@EzsXSyLGtey%1rBBXe
zr5>Kh&_nl1{X6MD@}NKHL4U-9e%OQls0aOF5BhgJ=y!O~z2x!2e_Y~lrf%2IcRlDz
z9%p_fkCXm@2hL+2^d~&%Klh-2!-M{e2fa{vej;(W;<4*T*$G!ZlK!pA%7bMG%S7ma
z%+HmOqQ~RUO?hsL2t6g~Mere#{vTwJ!L-kM4|<~qy}^~vm6ycV{0UcWlAg7HWzQuh
z{S~nRuTwC;l5%pjC-W;hs;}r=0cPkCyrTLVUE|9YqRgMwkFtNR7Be_E;T4NQg~L^)
zq-W)PD5JLw{YYhHby+p^d`Qx{`ZV#Co|XQKFfL2Frf05ZC0&!V%A~)7DV|u7)dN?v
zZuz<Hm3}6(auM(|w}q?3ZB$$~M8(RceU-bzO=7o=&I|#po~z_an_~b6PYgQ^qcBnj
zoCTPrpaN<lSO5nuSqkF7iWgLXuC|~8Y^#C_(3lG<z(y&k0L{Ok0_=AM6`-Rjr~rFR
zK?T?)3o5`qR!{*t#exdVF0?>np#_=>Eik9h0+?JDR_Ud-0_=ZQCt&OjuhP5*Hz|Ty
zub~$Rgfs4h9aGJdBM~PECfxZ!(ouhq*5QD(wAw**CpaG<IV*8g$0>SqmMgs*{B+u?
zus{g)=Z6>m{&L#8{&hBxBcj+HB>kcv$RayP%tMm&dq*si+@&?=0)Ujr=A_io=&c@-
zr0&Urlw_*7XV0E=Boo-|1hCRdEi`<RVT229cUTu{3G!Nnin<GR_Uu`QP|;}AA(WM7
z3PoB*CM=DW@K%~rD6I4>LVdK`L)~sdS+(-XG^O{(g%C>hMrz+7l*ueTU}QDrtXV_0
zX0(3>SCc+^^k^2uv;nJXO=ecw@YSt%%;gY6bxhU_ZDqFsXDzs~ri_NQa7<^+j+suU
z9JM%h?3lx7tTeM0qybw7(`c+T=agwK^5$wmJFuyCclbK3KQ2RoP}aRVYC*lTdRC`K
zy_+<%KRW$ZElTD5HWJQ8HbK1o7AMR(5QlWsaZtqKR_UCQ@GK@TNUF^r1RF&b5G35H
zqdR}S_e>iOh&q7s)(IC`d8-qPbK>a@!n3KoAgMNg5Oru~PPc*Rt`mQd={89`K-7Vp
zcXha6$y=RRoO5e$knzI-NwxWd@Nh3D2@>u)@dueP91waRzv^&NoRh?}@A!l8>@qJ%
zs?8q+>q-j<67D+j2bnn>5OrYd-Nan*=B-XF&J~h32+wl!f~4B~K_WH~B;0l44>D^w
zAnE|yyE@#w%3GaSoQptj5Z;5x3zBN{2bpRELBd@pb`X6WUx8w|3AV%RhXSK<e1_Fv
zOY@9KQ)E?VXRJ3lZ!o-83{uybV1#PY7h#*eS<Da-(S*1P2b7&c9Z#wyiGj-iC81dS
zk8htz7xA6EFm0H@%?z@s)Y-H++0^>5QHvJIrYIIa`{o<&Y)0#BTAgfW`mj;U8p)<O
z7XRy+H{IEc(b>#*vT5*P!-Y=}(<CUWBo=@B)cfvis&qCBoNVU!u;Kb5!zPTj`P)C;
z*^JfMEOfGI@nNIp!Ll}`vG}>aoOEYXO*ZrWEo}<e%pZoOrP`iOopxt4PG{p{Y14ez
zSS>Bp_Rcd;yR#XuvvIMsSw3v6mX>OJ?ew$mY%bQ>xLDe3A2wD?OSPqrzwXXvg3iXp
z(&qZGv07TH?a4Rab!RhCXX9dNtv+n5mX>P!_|+%e*-Rpvh5nWn0h@)xu(X!eyB~i|
zv7tjl$ojO&IvW>Do9@HLYH2O4ClB83&Zb6Z<6>!bK5VR(*3$Zm-yd>kGeu|PVrh*&
zY^;{n(t7N^7v0%JbT%%Q*6hQ^YH2O4Loa1KwB9yTbv7=RHs6Ph)zVs8KRUkOtu`ZZ
z&W(v$cHCU0jhmCxC7xrY5bO8;%pG)E4ipxpgHF{ztyY9Hlnra$L8s?HVLv+P3=Pz3
zGeR8x=@;BVXXHR(Ejnnu25PksAqKj>?+!XM2MXKJL7OyCt33#D^}{c?gU-r<!V+}Q
zc^as-^@aHNfy>=M!=etOt61IMH5guEZ+T-PE`Y7Z{$y8n1TNw(3bqtEJK{U{9W$P!
z0v8#u>A$AEr?bDOJ=xvYI~d+;2MkoF%Y?dn#JgcVfX>}TE0Hq^f^z@MWWfC>$)Feq
zRs6^e5Cdwkroy&VsVy}WD97~-9AB*27VF#|SrhB&84T~y<QkVA0lBteGT4b*dl9_H
zAYBitos)`tS8`Hl?WCqu+^{OnN~Kb!R4QdkrBXxK<()kcOb>=5c2SCgV6q06=hWpr
z@Dj}1I8cndbz0tHTxHJXjffjC65PllvP{!J2xVq!WkTsN%Jk#nS3=HxJ2O`!$z{$!
z2};b-N|dBWI!f4ijTLiwZT7OAJ9^rC+mn5XU6GBkoyo!Q6?Uk=f^-FLWbVYxJGOnV
zY=7RA+y$L)gg85451i|pt<T2Jm{%$4hW(r?`ubp~OYOMEgmg$;0j2e6rphgJ6AR2t
ztWmiBf+fe5vG%U+-mMX=6*NJE70~RJ5Qw(l4MDd;K($7YCV&nz<*MBZj;2@w$&@1;
z3XH-c!vYT9r-N6zfGhv!0iKTj;P{_V5rVf`!QlgS@X;>d$`5*gtEoXig3q^t!xw4}
zAvneboc>U?Scz3FLxB-u0UA<V)w`n~rqFM1vOpE;!)vKKV2f<@Tt`depulW2o+f<4
z?q7ik68utIcklLCSEm18tqC(8eyL41^ma6Qm(2qNv(}jPMW*-Ro_H}7H&QHu{9;|K
zKbdHUd(j-TF$s=&9e`QW1Z;|qMXhM0KgH>aS>Gc#g2@;>Olm-&JGR<oH3V3$gQX|r
z?OsA<v~S=ww^%L31xG-v>r2K4!yR^vzzFmgVk99HHz&F3k)mK{+jME%mCiX@l`>tL
zN|`QArG^4!Vu`>m*!upYZ1FE^OveIxz0`rT&8;q<9@5t4B;+JfDN_=aG9^(dy0S|K
zilDpcb`gq-&=4C?=R|hlPnFSBm(o->b|-tFt$FrR!E!XuMp#D@4M%-Bt=%BRG6~U_
zqy=bu55bEi1Qo7YKO~LHw!c8Vn|?&q)=bgFD#dcNk=WFp=%ve2?lo!`G*AXly$Ku=
z=vSuKa&xO3#HEJs3|}wqv}}^>N9`){w-F#6vbNVk<X!S@E$U8Z$b02Wpv7)+tF)k8
z4)&m}7)6tCl?%m9@(%LXQLDW?gN9{XyDb(=&N~zyFAiCFDzjC`E%)Qz_WhP`EVmD4
zY~b&N5k?eZ@_UB(Sh*G|oJQW&o{XXYb@cW1#M*oD*4uw`M=UWf`__Vtu|%RTk;Y6Q
zEZ)Z{I+Q6o!_|mnqCz!(0EY%X06`8|zN~)R5QjlKY@t!`eW2~L(1^pU66ykah3x`*
zS)huaWuV{(akBF(oCRM5e(Zn&sJC3cSK`V4h!Peb|M1Lj*LG}(Zjtl`NuMpBUzL;r
z<T1kxn7t5<#$J3m5hAa#h+mVrZ<o)R$Q_sAoha|On(uX4ly{@`miPO30~S(I{f0yk
zqE&|9Ma^H#0Wrrok`n_Z>|q&)9~rI#1~Ot4W1tSGYwYj_cq+vAMr92~xuh|n)0khU
zVcc=hR6FMI&dGLB|HVg$ub{uu<W`99wVYnbZ{$@jpT)Qx%=$7#D*yYzaUsr?h#G_c
z%9x#Eu(1>r8<3Yp{t@Ljic>D1K~$BFzFZu-hvAI;PSMJxp^+7>n@BKL$qG!*D^}<;
zuc7-)fDjvSt$8kDM6AFz<dZyW5G(n}ggan|ci7q>0B8s?juSI*7>B=#iml;sVyh(^
z<mO&L@KR;9CfaBIm(9kuVN3A;XVKKCGwPUW?l<%ILil`Y;5@o)ksN=}(3U{EIQ8{=
z&wjXmTg5|ng~jYykDr7G7T-psZSea_ME1}t5!qEXB92D936X=wPDE_XiQ5s!Anr$`
zt(`_89u=}IgZ|EbxVF19(RWK<JQ=xcXJ@QO@*3FxhX+m`df8aXLiNud<vZ>6lLWu|
z2VQ;Y)gyBzRtz0P{Tn}gg!O+Ok$r;o-+)M8c0D4`l)DglnU6C~)`63C*84U@4w$zi
zQXl&esh@ih-SnaAG*V(Ai|20*m{1P3hng{nksuM0Dy1l!`V!l3;bh_#*~6LJ9*oSg
zcn2n&9pa1jkHbjue_>gB#xROVTcNHlg#W(=40!h-MtHWNF3n63a{X|9Jl@?IQ{6Gr
g(%jq{$r_g&F;;iS<BKK42TP89Nyoa7{&Z^Kza0A5{r~^~

literal 0
HcmV?d00001

diff --git a/pandas/io/tests/test_excel.py b/pandas/io/tests/test_excel.py
index 1ac4d4e31ed10..f5db9d0c016af 100644
--- a/pandas/io/tests/test_excel.py
+++ b/pandas/io/tests/test_excel.py
@@ -18,7 +18,8 @@
 import pandas.io.parsers as parsers
 from pandas.io.parsers import (read_csv, read_table, read_fwf,
                                 TextParser, TextFileReader)
-from pandas.io.excel import ExcelFile, ExcelWriter, read_excel
+from pandas.io.excel import ExcelFile, ExcelWriter, read_excel,\
+    excel_value_to_python_value
 from pandas.util.testing import (assert_almost_equal,
                                  assert_series_equal,
                                  network,
@@ -260,6 +261,32 @@ def test_excel_table(self):
         tm.assert_frame_equal(df4, df.ix[:-1])
         tm.assert_frame_equal(df4, df5)
 
+    def test_excel_value_to_python_value(self):
+        _skip_if_no_xlrd()
+
+        pth = os.path.join(self.dirpath, 'types.xls')
+        xls = ExcelFile(pth)        
+        book = xls.book
+        sheet = book.sheet_by_index(0)
+        
+        cell = sheet.cell(0, 0)
+        self.assertEquals(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), 'date')
+
+        cell = sheet.cell(0, 1)
+        self.assertEquals(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), datetime(year=2013, month=4, day=1))  
+        
+        cell = sheet.cell(1, 1)
+        self.assertEquals(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), True)
+        
+        cell = sheet.cell(2, 1)
+        self.assertEquals(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), 1)
+
+        cell = sheet.cell(3, 1)
+        self.assertEquals(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), 1.1)
+        
+        cell = sheet.cell(4, 1)
+        self.assertIs(excel_value_to_python_value(value=cell.value, typ=cell.ctype, datemode=book.datemode), np.nan) #We need to use is here because value is NaN
+        
     def test_excel_read_buffer(self):
         _skip_if_no_xlrd()
         _skip_if_no_openpyxl()