Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Billy Schwartz <wkschwartz@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: import excel - requiring restart occasionally |

Date |
Wed, 2 Nov 2011 20:31:35 -0400 |

describe" eventually lead to increasing numbers of error 603's when running the import command. Restarting Stata seems to fix it. Facts: I have 947 Excel workbooks, both .xls and .xlsx, averaging just under 3 sheets per book, and I have to organize them. As a first step I've written a Stata script to create a table of the sheet names in each workbook. It loops through the *.xls* file names in my folder and runs "import excel, describe" on each, putting the contents returned in r() into a data set that accumulates. I open up Stata and run the script, and it works as intended with 14 error 603s, which I catch and handle. (So you don't have to look it up, error 603 means the file could not be opened even though it was found. I don't know why these error 603's occur since Excel opens the file just fine, but I'm comfortable with a 1% error rate.) During the debugging process, I've had to run the script repeatedly in a given instance of Stata. After three to five runs, the number of error 603's I get goes from 14 to a couple hundred. One or two more runs gives me an error 603 on each iteration of the loop, leaving my dataset empty. After restarting Stata, the problem goes away and I'm back to 14 error 603's. I'm doing all this on a server, so I've tried reading the data from a different server over the network and locally on this server. Same pattern both times. Question: Has anyone had a similar problem? Can anyone replicate this? The code below the fold is a skeleton of the algorithm I'm using in case anyone wants to try. If everyone's out of ideas, I guess I'll submit this to Stata as a bug report. Technical details in case it matters: My copy of Stata 12/MP (4-core license, born on 13oct2011) runs on Windows Server Enterprise 2007 SP1 (64bit, 32GB RAM, four 4-core Xeons @ 1.87Ghz). *! stata version 12 clear local directory "." //replace as appropriate local files: dir "`directory'" files "*.xls*", respectcase generate book_name = "" generate int error = . generate sheet_name = "" generate sheet_range = "" foreach file of local files { capture import excel using "`directory'/`file'", describe if c(rc) == 603 { set obs `=c(N)+1' replace book_name = "`file'" in `c(N)' replace error = c(rc) in `c(N)' continue } else error c(rc) forvalues i = 1/`r(N_worksheet)' { set obs `=c(N)+1' replace book_name = "`file'" in `c(N)' replace error = c(rc) in `c(N)' replace sheet_name = r(worksheet_`i') in `c(N)' replace sheet_range = "`r(range_`i')'" in `c(N)' //range may be missing } } count if error //number of errors *end * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**Re: st: How to use tobit where the low bounds are not fixed for each observation?** - Next by Date:
**st: log files** - Previous by thread:
**st: Insert new observation by group** - Next by thread:
**Re: st: import excel - requiring restart occasionally** - Index(es):