Re: [Rd] read.table: wrong error message? (PR#10592)
[EMAIL PROTECTED] wrote: --Apple-Mail-44--797532055 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit I believe read.table may report misleading errors. In this example, where a header line in a file has an incorrect number of row names (28 instead of 29), I get the error message duplicate row.names are not allowed. However, I cannot not find any duplicate row names. Fixing the header line by adding an extra row name, however, avoids the error. The behavior is confusing - I would expect a different error message, even if I should have quoted some row names. Arguably, read.table() is too smart for its own good at times, but this _is_ documented behaviour: If 'row.names' is not specified and the header line has one less entry than the number of columns, the first column is taken to be the row names. This allows data frames to be read in from the format in which they are printed. If 'row.names' is specified and does not refer to the first column, that column is discarded from such files. So, if your COLUMN (sic!) name count is off by one, read.table() takes the first variable as row.names, and when this is a 0/1 variable, you will have duplicate names. I am attaching a sample file that reproduces the problem with R -- vanilla, and then: read.table(m2). I get it in 2.5.1 and 2.6.0. PS.: R version 2.6.0 (2007-10-03) ... x - read.table(m2, header=T) Error in read.table(m2, header = T) : duplicate 'row.names' are not allowed traceback() 2: stop(duplicate 'row.names' are not allowed) 1: read.table(m2, header = T) --Apple-Mail-44--797532055 Content-Disposition: attachment; filename=m2 Content-Type: application/octet-stream; x-unix-mode=0644; name=m2 Content-Transfer-Encoding: 7bit primed rule role dist starttime target.utt rule.freq primeperiod.length dialogue.length dialogue.id words.repeated words.repeated.prop head.repeated head.freq head.pos prime.gaze target.gaze eyecontact familiar convseq length doc.score friend task.familiar same.specrule derivation pathlen distituent 0 vp---vbg-vp i 2 5.67 4 False 172 70 261.5322 1 0 0/7 True na None None None 1 0 1 7 135 - - 1 None - - 0 vp---to-vp r 1 6.03 4 False 758 13 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 vp---to-vp i 2 6.03 4 False 758 70 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 s---cc-s i 1 11.3 5 False 813 83 261.5322 1 0 2/9 True na None None None 1 0 1 9 135 - - 1 None - - 0 s---cc-s i 3 11.3 5 False 813 70 261.5322 1 0 0/9 True na None None None 1 0 1 9 135 - - 1 None - - 0 s---advp-s i 3 12.71 5 False 440 70 261.5322 1 0 0/8 True na None None None 1 0 1 8 135 - - 1 None - - 1 vp---ber-vp i 1 12.98 5 False 406 83 261.5322 1 0 2/7 True na None None None 1 0 1 7 135 - - 1 None - - 1 vp---vbg-vp i 1 13.11 5 False 172 83 261.5322 1 0 3/7 True na None None None 1 0 1 7 135 - - 1 None - - 1 vp---to-vp i 1 13.52 5 False 758 83 261.5322 1 0 2/6 True na None None None 1 0 1 6 135 - - 1 None - - 1 pp---ql-rp-pp i 1 14.12 5 False 282 83 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - 0 pp---ql-rp-pp r 2 14.12 5 False 282 13 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - 0 ap---ap r 2 18.75 5 False 120 13 261.5322 1 0 0/1 True na None None None 1 0 1 1 135 - - 1 None - - 0 pp---pp-pp i 3 22.7 6 False 429 13 261.5322 1 0 0/4 True na None None None 1 0 1 4 135 - - 1 None - - 1 pp---ql-rp r 2 22.7 6 False 505 83 261.5322 1 0 1/2 True na None None None 1 0 1 2 135 - - 1 None - - 0 pp---rp-pp r 2 23.54 6 False 1124 83 261.5322 1 0 0/2 True na None None None 1 0 1 2 135 - - 1 None - - 0 pp---pp-cc-rb-pp i 2 25.27 7 False 64 200 261.5322 1 0 1/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 pp---pp-cc-rb-pp r 4 25.27 7 False 64 13 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 1 pp---ql-rp-pp i 2 25.99 7 False 282 200 261.5322 1 0 2/3 True na None None None 1 0 1 3 135 - - 1 None - - 1 pp---ql-rp-pp i 3 25.99 7 False 282 83 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - --Apple-Mail-44--797532055-- __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] read.table: wrong error message? (PR#10592)
On Sun, 20 Jan 2008, [EMAIL PROTECTED] wrote: --Apple-Mail-44--797532055 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit I believe read.table may report misleading errors. In this example, where a header line in a file has an incorrect number of row names (28 instead of 29), I get the error message duplicate row.names are not allowed. The first column of your file repeats 0 and 1, and your specification asked for the first column to be taken as the row names: from the help file If 'row.names' is not specified and the header line has one less entry than the number of columns, the first column is taken to be the row names. However, I cannot not find any duplicate row names. Fixing the header line by adding an extra row name, however, avoids the error. An extra *column* name? The behaviour of header=TRUE depends on the number of fields on the header line: see the quote above. The behavior is confusing - I would expect a different error message, even if I should have quoted some row names. It may have confused you, but it has been in R/S for 20 years and is described in the help, the 'R Data Import/Export Manual', most books on R/S It is fortunate that you did get an error message: if you give an input format different from what you intended it can easily be valid but not that you intended. If you want to be safer, supply the argument 'row.names'. I am attaching a sample file that reproduces the problem with R -- vanilla, and then: read.table(m2). I get it in 2.5.1 and 2.6.0. Also 2.6,1 and R-devel, as it is the documented behaviour. PS.: R version 2.6.0 (2007-10-03) ... x - read.table(m2, header=T) Error in read.table(m2, header = T) : duplicate 'row.names' are not allowed traceback() 2: stop(duplicate 'row.names' are not allowed) 1: read.table(m2, header = T) --Apple-Mail-44--797532055 Content-Disposition: attachment; filename=m2 Content-Type: application/octet-stream; x-unix-mode=0644; name=m2 Content-Transfer-Encoding: 7bit primed rule role dist starttime target.utt rule.freq primeperiod.length dialogue.length dialogue.id words.repeated words.repeated.prop head.repeated head.freq head.pos prime.gaze target.gaze eyecontact familiar convseq length doc.score friend task.familiar same.specrule derivation pathlen distituent 0 vp---vbg-vp i 2 5.67 4 False 172 70 261.5322 1 0 0/7 True na None None None 1 0 1 7 135 - - 1 None - - 0 vp---to-vp r 1 6.03 4 False 758 13 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 vp---to-vp i 2 6.03 4 False 758 70 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 s---cc-s i 1 11.3 5 False 813 83 261.5322 1 0 2/9 True na None None None 1 0 1 9 135 - - 1 None - - 0 s---cc-s i 3 11.3 5 False 813 70 261.5322 1 0 0/9 True na None None None 1 0 1 9 135 - - 1 None - - 0 s---advp-s i 3 12.71 5 False 440 70 261.5322 1 0 0/8 True na None None None 1 0 1 8 135 - - 1 None - - 1 vp---ber-vp i 1 12.98 5 False 406 83 261.5322 1 0 2/7 True na None None None 1 0 1 7 135 - - 1 None - - 1 vp---vbg-vp i 1 13.11 5 False 172 83 261.5322 1 0 3/7 True na None None None 1 0 1 7 135 - - 1 None - - 1 vp---to-vp i 1 13.52 5 False 758 83 261.5322 1 0 2/6 True na None None None 1 0 1 6 135 - - 1 None - - 1 pp---ql-rp-pp i 1 14.12 5 False 282 83 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - 0 pp---ql-rp-pp r 2 14.12 5 False 282 13 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - 0 ap---ap r 2 18.75 5 False 120 13 261.5322 1 0 0/1 True na None None None 1 0 1 1 135 - - 1 None - - 0 pp---pp-pp i 3 22.7 6 False 429 13 261.5322 1 0 0/4 True na None None None 1 0 1 4 135 - - 1 None - - 1 pp---ql-rp r 2 22.7 6 False 505 83 261.5322 1 0 1/2 True na None None None 1 0 1 2 135 - - 1 None - - 0 pp---rp-pp r 2 23.54 6 False 1124 83 261.5322 1 0 0/2 True na None None None 1 0 1 2 135 - - 1 None - - 0 pp---pp-cc-rb-pp i 2 25.27 7 False 64 200 261.5322 1 0 1/6 True na None None None 1 0 1 6 135 - - 1 None - - 0 pp---pp-cc-rb-pp r 4 25.27 7 False 64 13 261.5322 1 0 0/6 True na None None None 1 0 1 6 135 - - 1 None - - 1 pp---ql-rp-pp i 2 25.99 7 False 282 200 261.5322 1 0 2/3 True na None None None 1 0 1 3 135 - - 1 None - - 1 pp---ql-rp-pp i 3 25.99 7 False 282 83 261.5322 1 0 0/3 True na None None None 1 0 1 3 135 - - 1 None - - --Apple-Mail-44--797532055-- __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595
Re: [Rd] read.table: wrong error message? (PR#10592)
--Apple-Mail-1--729254567 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit On 21 Jan 2008, at 11:38, Prof Brian Ripley wrote: The first column of your file repeats 0 and 1, and your specification asked for the first column to be taken as the row names: from the help file If 'row.names' is not specified and the header line has one less entry than the number of columns, the first column is taken to be the row names. I understand now. Thanks for the explanation. I would have figured it out myself had I paid attention to the fact that the error message reports row names rather than column names, or had the error message actually contained the conflicting names (either 0 or 1) or the full set of read names. -- David Reitter ICCS/HCRC, Informatics, University of Edinburgh http://www.david-reitter.com --Apple-Mail-1--729254567 Content-Disposition: attachment; filename=smime.p7s Content-Type: application/pkcs7-signature; name=smime.p7s Content-Transfer-Encoding: base64 MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIFxDCCAn0w ggHmoAMCAQICEEfSZyInpN9bVdFPsdICbiIwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkEx JTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQ ZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MTIwOTIzMDM0OFoXDTA4MTIwODIzMDM0 OFowYjEQMA4GA1UEBBMHUmVpdHRlcjEOMAwGA1UEKhMFRGF2aWQxFjAUBgNVBAMTDURhdmlkIFJl aXR0ZXIxJjAkBgkqhkiG9w0BCQEWF2RhdmlkLnJlaXR0ZXJAZ21haWwuY29tMIGfMA0GCSqGSIb3 DQEBAQUAA4GNADCBiQKBgQDIS4kZGULD+CxprkxDnEccAnZ2GQQxfn55aiAPoy5kJ+uLhelaZP4p lgx3Vq8xYK1bjsLepGYp8qzVlHsyQPb6OfvRzrUXHxlJrHME0+Lblx37PBLxGDxc5W8qfZFJpej8 Dk3LUMga8laR9CRghbALoTV3UThUSMqk1cVwTY99KQIDAQABozQwMjAiBgNVHREEGzAZgRdkYXZp ZC5yZWl0dGVyQGdtYWlsLmNvbTAMBgNVHRMBAf8EAjAAMA0GCSqGSIb3DQEBBQUAA4GBAKU8XIS6 R/R3Hg6ae5p0i4iUJURq+sDVMX9RGB3Ge5V8oKKglIQGaY03leem65pqICL6mdgX/Px21JoaKVAA 8XITLFVEFHL6BfZ/eePnzNi93rnaNe7hLUwLdGztdYmqN8zF0DOCybL1M1JDB3cTr0XwHJwxziSf oEeENGNETykjMIIDPzCCAqigAwIBAgIBDTANBgkqhkiG9w0BAQUFADCB0TELMAkGA1UEBhMCWkEx FTATBgNVBAgTDFdlc3Rlcm4gQ2FwZTESMBAGA1UEBxMJQ2FwZSBUb3duMRowGAYDVQQKExFUaGF3 dGUgQ29uc3VsdGluZzEoMCYGA1UECxMfQ2VydGlmaWNhdGlvbiBTZXJ2aWNlcyBEaXZpc2lvbjEk MCIGA1UEAxMbVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIENBMSswKQYJKoZIhvcNAQkBFhxwZXJz b25hbC1mcmVlbWFpbEB0aGF3dGUuY29tMB4XDTAzMDcxNzAwMDAwMFoXDTEzMDcxNjIzNTk1OVow YjELMAkGA1UEBhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAq BgNVBAMTI1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMIGfMA0GCSqGSIb3DQEB AQUAA4GNADCBiQKBgQDEpjxVc1X7TrnKmVoeaMB1BHCd3+n/ox7svc31W/Iadr1/DDph8r9RzgHU 5VAKMNcCY1osiRVwjt3J8CuFWqo/cVbLrzwLB+fxH5E2JCoTzyvV84J3PQO+K/67GD4Hv0CAAmTX p6a7n2XRxSpUhQ9IBH+nttE8YQRAHmQZcmC3+wIDAQABo4GUMIGRMBIGA1UdEwEB/wQIMAYBAf8C AQAwQwYDVR0fBDwwOjA4oDagNIYyaHR0cDovL2NybC50aGF3dGUuY29tL1RoYXd0ZVBlcnNvbmFs RnJlZW1haWxDQS5jcmwwCwYDVR0PBAQDAgEGMCkGA1UdEQQiMCCkHjAcMRowGAYDVQQDExFQcml2 YXRlTGFiZWwyLTEzODANBgkqhkiG9w0BAQUFAAOBgQBIjNFQg+oLLswNo2asZw9/r6y+whehQ5aU nX9MIbj4Nh+qLZ82L8D0HFAgk3A8/a3hYWLD2ToZfoSxmRsAxRoLgnSeJVCUYsfbJ3FXJY3dqZw5 jowgT2Vfldr394fWxghOrvbqNOUQGls1TXfjViF4gtwhGTXeJLHTHUb/XV9lTzGCAo8wggKLAgEB MHYwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4x LDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBAhBH0mciJ6TfW1XR T7HSAm4iMAkGBSsOAwIaBQCgggFvMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcN AQkFMQ8XDTA4MDEyMTExNTA0NVowIwYJKoZIhvcNAQkEMRYEFJN8Qc4lkZZVU7GKZnk9AnkCkaj4 MIGFBgkrBgEEAYI3EAQxeDB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3Vs dGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWlu ZyBDQQIQR9JnIiek31tV0U+x0gJuIjCBhwYLKoZIhvcNAQkQAgsxeKB2MGIxCzAJBgNVBAYTAlpB MSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUg UGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQR9JnIiek31tV0U+x0gJuIjANBgkqhkiG9w0B AQEFAASBgJaMRB2ro5JDA0vHTeC8kLMSU6sI1OF4hew1+u3BYe5iIEgoTM7rh8ICl5v7v2JUqCme vIq+bG3Hm81MTr9f3WNumrDyCTJHIthKyn8yJkk8mXT7rGSwR3378iFu0J8IZZ6Fa/895enf9FLs 7119hVIlS0G4t4/d1R8hbCevrUtQ --Apple-Mail-1--729254567-- __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] read.table: wrong error message? (PR#10592)
On Mon, 21 Jan 2008, David Reitter wrote: On 21 Jan 2008, at 11:38, Prof Brian Ripley wrote: The first column of your file repeats 0 and 1, and your specification asked for the first column to be taken as the row names: from the help file If 'row.names' is not specified and the header line has one less entry than the number of columns, the first column is taken to be the row names. I understand now. Thanks for the explanation. I would have figured it out myself had I paid attention to the fact that the error message reports row names rather than column names, or had the error message actually contained the conflicting names (either 0 or 1) or the full set of read names. We can fairly easily do something along those lines: I've added code to R-devel to do library(MASS) row.names(hills)[35] - Lomonds Error in `row.names-.data.frame`(`*tmp*`, value = c(Greenmantle, Carnethy, : duplicate 'row.names' are not allowed In addition: Warning message: non-unique value when setting 'row.names': ‘Lomonds’ -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
