Improve bug reporting documentation, and mention pspp-security list.
[pspp] / tests / language / stats / examine.at
index 0acbf5f33035030953ee7a1575ad3d58da8a435a..6ccfac555a7d150a2d4e5f95d806c65bb3e1f361 100644 (file)
@@ -1,3 +1,19 @@
+dnl PSPP - a program for statistical analysis.
+dnl Copyright (C) 2017 Free Software Foundation, Inc.
+dnl 
+dnl This program is free software: you can redistribute it and/or modify
+dnl it under the terms of the GNU General Public License as published by
+dnl the Free Software Foundation, either version 3 of the License, or
+dnl (at your option) any later version.
+dnl 
+dnl This program is distributed in the hope that it will be useful,
+dnl but WITHOUT ANY WARRANTY; without even the implied warranty of
+dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+dnl GNU General Public License for more details.
+dnl 
+dnl You should have received a copy of the GNU General Public License
+dnl along with this program.  If not, see <http://www.gnu.org/licenses/>.
+dnl
 AT_BANNER([EXAMINE])
 
 AT_SETUP([EXAMINE])
 AT_BANNER([EXAMINE])
 
 AT_SETUP([EXAMINE])
@@ -36,10 +52,11 @@ EXAMINE
        /STATISTICS descriptives extreme(3)
        .
 ])
        /STATISTICS descriptives extreme(3)
        .
 ])
-AT_CHECK([pspp -o pspp.csv examine.sps])
+
+
 dnl In the following data, only the extreme values have been checked.
 dnl The descriptives have been blindly pasted.
 dnl In the following data, only the extreme values have been checked.
 dnl The descriptives have been blindly pasted.
-AT_CHECK([cat pspp.csv], [0], [dnl
+AT_CHECK([pspp -O format=csv examine.sps], [0], [dnl
 Table: Reading free-form data from INLINE.
 Variable,Format
 QUALITY,F8.0
 Table: Reading free-form data from INLINE.
 Variable,Format
 QUALITY,F8.0
@@ -77,8 +94,8 @@ Breaking Strain,Highest,1,12,7.00
 ,,2,16,6.00
 ,,3,14,5.00
 ,Lowest,1,3,1.00
 ,,2,16,6.00
 ,,3,14,5.00
 ,Lowest,1,3,1.00
-,,2,3,1.00
-,,3,4,1.00
+,,2,4,1.00
+,,3,2,2.00
 
 Table: Descriptives
 ,,,Statistic,Std. Error
 
 Table: Descriptives
 ,,,Statistic,Std. Error
@@ -110,11 +127,11 @@ Breaking Strain,Aspeger,Highest,1,6,4.00
 ,,,2,5,4.00
 ,,,3,1,3.00
 ,,Lowest,1,3,1.00
 ,,,2,5,4.00
 ,,,3,1,3.00
 ,,Lowest,1,3,1.00
-,,,2,3,1.00
-,,,3,4,1.00
+,,,2,4,1.00
+,,,3,2,2.00
 ,Bloggs,Highest,1,7,5.00
 ,,,2,9,4.00
 ,Bloggs,Highest,1,7,5.00
 ,,,2,9,4.00
-,,,3,9,4.00
+,,,3,11,3.00
 ,,Lowest,1,8,2.00
 ,,,2,10,2.00
 ,,,3,11,3.00
 ,,Lowest,1,8,2.00
 ,,,2,10,2.00
 ,,,3,11,3.00
@@ -123,7 +140,7 @@ Breaking Strain,Aspeger,Highest,1,6,4.00
 ,,,3,14,5.00
 ,,Lowest,1,15,3.00
 ,,,2,13,4.00
 ,,,3,14,5.00
 ,,Lowest,1,15,3.00
 ,,,2,13,4.00
-,,,3,13,4.00
+,,,3,14,5.00
 
 Table: Descriptives
 ,Manufacturer,,,Statistic,Std. Error
 
 Table: Descriptives
 ,Manufacturer,,,Statistic,Std. Error
@@ -167,6 +184,7 @@ Breaking Strain,Aspeger,Mean,,2.25,.45
 ,,Skewness,,.30,.75
 ,,Kurtosis,,.15,1.48
 ])
 ,,Skewness,,.30,.75
 ,,Kurtosis,,.15,1.48
 ])
+
 AT_CLEANUP
 
 AT_SETUP([EXAMINE -- extremes])
 AT_CLEANUP
 
 AT_SETUP([EXAMINE -- extremes])
@@ -202,8 +220,8 @@ examine v1
  /statistics=extreme(6)
  .
 ])
  /statistics=extreme(6)
  .
 ])
-AT_CHECK([pspp -o pspp.csv examine.sps])
-AT_CHECK([cat pspp.csv], [0], [dnl
+
+AT_CHECK([pspp -O format=csv examine.sps], [0],[dnl
 Table: Case Processing Summary
 ,Cases,,,,,
 ,Valid,,Missing,,Total,
 Table: Case Processing Summary
 ,Cases,,,,,
 ,Valid,,Missing,,Total,
@@ -215,16 +233,17 @@ Table: Extreme Values
 V1,Highest,1,21,20.00
 ,,2,20,19.00
 ,,3,19,18.00
 V1,Highest,1,21,20.00
 ,,2,20,19.00
 ,,3,19,18.00
-,,4,19,18.00
-,,5,18,17.00
-,,6,17,16.00
+,,4,18,17.00
+,,5,17,16.00
+,,6,16,15.00
 ,Lowest,1,1,1.00
 ,,2,2,2.00
 ,,3,3,3.00
 ,Lowest,1,1,1.00
 ,,2,2,2.00
 ,,3,3,3.00
-,,4,3,3.00
-,,5,4,3.00
-,,6,5,4.00
+,,4,4,3.00
+,,5,5,4.00
+,,6,6,5.00
 ])
 ])
+
 AT_CLEANUP
 
 
 AT_CLEANUP
 
 
@@ -612,7 +631,12 @@ BEGIN DATA.
 .
 END DATA.
 
 .
 END DATA.
 
-EXAMINE /x PLOT=HISTOGRAM.
+EXAMINE /x 
+       PLOT=HISTOGRAM BOXPLOT NPPLOT SPREADLEVEL(1) ALL
+       /ID=x
+        /STATISTICS = DESCRIPTIVES EXTREME (5) ALL
+       /PERCENTILE=AEMPIRICAL
+       .
 ])
 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 dnl Ignore output -- this is just a no-crash check.
 ])
 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 dnl Ignore output -- this is just a no-crash check.
@@ -870,4 +894,240 @@ x,. (missing),Highest,1,31,4004
 ])
 
 
 ])
 
 
-AT_CLEANUP 
\ No newline at end of file
+AT_CLEANUP 
+
+
+dnl Run a test of the basic STATISTICS using a "real"
+dnl dataset and comparing with "real" results kindly
+dnl provided by Olaf Nöhring
+AT_SETUP([EXAMINE -- sample unweighted])
+
+AT_DATA([sample.sps], [dnl
+set format = F22.4.
+DATA LIST notable LIST /X *
+BEGIN DATA.
+461.19000000
+466.38000000
+479.46000000
+480.10000000
+483.43000000
+488.30000000
+489.00000000
+491.62000000
+505.62000000
+511.30000000
+521.53000000
+526.70000000
+528.25000000
+538.70000000
+540.22000000
+540.58000000
+546.10000000
+548.17000000
+553.99000000
+566.21000000
+575.90000000
+584.38000000
+593.40000000
+357.05000000
+359.73000000
+360.48000000
+373.98000000
+374.13000000
+381.45000000
+383.72000000
+390.00000000
+400.34000000
+415.32000000
+415.91000000
+418.30000000
+421.03000000
+422.43000000
+426.93000000
+433.25000000
+436.89000000
+445.33000000
+446.33000000
+446.55000000
+456.44000000
+689.49000000
+691.92000000
+695.00000000
+695.36000000
+698.21000000
+699.46000000
+706.61000000
+710.69000000
+715.82000000
+715.82000000
+741.39000000
+752.27000000
+756.73000000
+757.74000000
+759.57000000
+796.07000000
+813.78000000
+817.25000000
+825.48000000
+831.28000000
+849.24000000
+890.00000000
+894.78000000
+935.65000000
+935.90000000
+945.90000000
+1012.8600000
+1022.6000000
+1061.8100000
+1063.5000000
+1077.2300000
+1151.6300000
+1355.2800000
+598.88000000
+606.91000000
+621.60000000
+624.80000000
+636.13000000
+637.38000000
+640.32000000
+649.35000000
+656.51000000
+662.55000000
+664.69000000
+106.22000000
+132.24000000
+174.76000000
+204.85000000
+264.93000000
+264.99000000
+269.84000000
+325.12000000
+331.67000000
+337.26000000
+347.68000000
+354.91000000
+END DATA.
+
+EXAMINE
+       x
+       /STATISTICS=DESCRIPTIVES
+       .
+])
+
+AT_CHECK([pspp -O format=csv sample.sps], [0], [dnl
+Table: Case Processing Summary
+,Cases,,,,,
+,Valid,,Missing,,Total,
+,N,Percent,N,Percent,N,Percent
+X,100,100%,0,0%,100,100%
+
+Table: Descriptives
+,,,Statistic,Std. Error
+X,Mean,,587.6603,23.2665
+,95% Confidence Interval for Mean,Lower Bound,541.4946,
+,,Upper Bound,633.8260,
+,5% Trimmed Mean,,579.7064,
+,Median,,547.1350,
+,Variance,,54132.8466,
+,Std. Deviation,,232.6647,
+,Minimum,,106.2200,
+,Maximum,,1355.2800,
+,Range,,1249.0600,
+,Interquartile Range,,293.1575,
+,Skewness,,.6331,.2414
+,Kurtosis,,.5300,.4783
+])
+
+AT_CLEANUP 
+
+
+
+dnl Test for a crash which happened on bad input syntax
+AT_SETUP([EXAMINE -- Empty Parentheses])
+
+AT_DATA([examine-empty-parens.sps], [dnl
+DATA LIST notable LIST /X *
+BEGIN DATA.
+2
+3
+END DATA.
+
+
+EXAMINE
+       x
+       /PLOT = SPREADLEVEL()
+       .
+])
+
+AT_CHECK([pspp -o pspp.csv examine-empty-parens.sps], [1], [ignore])
+
+AT_CLEANUP 
+
+
+
+
+dnl Test for another crash which happened on bad input syntax
+AT_SETUP([EXAMINE -- Bad variable])
+
+AT_DATA([examine-bad-variable.sps], [dnl
+data list list /h * g *.
+begin data.
+3 1
+4 1
+5 2
+end data.
+
+EXAMINE
+        /VARIABLES/ h
+        BY  g
+        .
+])
+
+AT_CHECK([pspp -o pspp.csv examine-bad-variable.sps], [1], [ignore])
+
+AT_CLEANUP 
+
+
+
+dnl Test for yet another crash. This time for extremes vs. missing weight values.\0
+AT_SETUP([EXAMINE -- Extremes vs. Missing Weights])
+
+AT_DATA([examine-missing-weights.sps], [dnl
+data list notable list /h * g *.
+begin data.
+3 1
+4 .
+5 1
+2 1
+end data.
+
+WEIGHT BY g.
+
+EXAMINE h
+       /STATISTICS extreme(3)
+       .
+])
+
+AT_CHECK([pspp -O format=csv  examine-missing-weights.sps], [0], [dnl
+"examine-missing-weights.sps:13: warning: EXAMINE: At least one case in the data file had a weight value that was user-missing, system-missing, zero, or negative.  These case(s) were ignored."
+
+Table: Case Processing Summary
+,Cases,,,,,
+,Valid,,Missing,,Total,
+,N,Percent,N,Percent,N,Percent
+h,3.00,100%,.00,0%,3.00,100%
+
+Table: Extreme Values
+,,,Case Number,Value
+h,Highest,1,3,5.00
+,,2,2,4.00
+,,3,1,3.00
+,Lowest,1,4,2.00
+,,2,1,3.00
+,,3,2,4.00
+])
+
+AT_CLEANUP 
+
+
+