<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">That's good. The warnings are because
      of that the snp name in the second line of TEST10x15.geno file is
      the same as in the first one. Fixed in
      <a class="moz-txt-link-freetext" href="https://app.box.com/s/6irjkikm3mukwvtk1wgw">https://app.box.com/s/6irjkikm3mukwvtk1wgw</a> version. Can you
      download it and run once again?<br>
      <br>
      BTW1: The versions of GenABEL and DatABEL here are old (I
      generated this example a week ago). That's why there are some
      warnings on other functions.<br>
      BTW2: Remove the files TEST10x15_T.geno.dose.fvd  and
      TEST10x15_T.geno.dose.fvi after each run. Otherwise you get an
      error message (File 'TEST10x15_T.geno.dose' already exists.).<br>
      <br>
      best,<br>
      Maksim<br>
      <br>
      <br>
      On 25/11/2013 19:28, Yury Aulchenko wrote:<br>
    </div>
    <blockquote
      cite="mid:F677E706-1CC0-406E-AA7D-945676E1CB72@gmail.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html;
        charset=ISO-8859-1">
      after renaming out-file-name I get (finished with warnings):
      <div><br>
      </div>
      <div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">> x <- impute2databel(geno="TEST10x15.geno",
          sample="impute.sample5", out="TEST10x15_T1.geno",
          makeprob=FALSE, old=TRUE)</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Loading required package: DatABEL</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">DatABEL v.0.9-4 (March 12, 2013) loaded</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0); min-height: 20px;"><br>
        </div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Options in effect: </div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --infile    = TEST10x15.geno</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --outfile   = ./tmp300070</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --skiprows  = OFF</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --skipcols  = 5</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --cnrow     = OFF</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --rncol     = ON, using column 2 of 'TEST10x15.geno'</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --transpose = ON</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --Rmatrix   = OFF</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);"><span class="Apple-tab-span" style="white-space:pre"> </span>
          --nanString = NA</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Number of lines in source file is 10</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Number of words in source file is 20</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">skiprows = 0</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">cnrow = 0</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">skipcols = 5</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">rncol = 2</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Rmatrix = 0</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">numWords = 20</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Creating file with numRows = 10</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Creating file with numColumns = 15</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">text2fvf finished.</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Loss of precision / loss of data during conversion from
          DOUBLE to FLOAT.</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Futher conversion warnings omitted.</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Read 4 items</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Read 4 items</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Read 4 items</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Read 20 items</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">Warning messages:</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">1: In uninames(.Object@data) :</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">  uninames: some column names are not unique; use
          set_dimnames/get_dimnames for non-unique row/col names</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">2: In uninames(x@data) :</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">  uninames: some column names are not unique; use
          set_dimnames/get_dimnames for non-unique row/col names</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">3: In uninames(x@data) :</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">  uninames: some column names are not unique; use
          set_dimnames/get_dimnames for non-unique row/col names</div>
        <div style="margin: 0px; font-size: 18px; font-family: 'Andale
          Mono'; color: rgb(41, 249, 20); background-color: rgb(0, 0,
          0);">> </div>
        <div><br>
        </div>
        <div><br>
        </div>
        <div>
          <div>On Nov 25, 2013, at 13:23 PM, L.C. Karssen <<a
              moz-do-not-send="true" href="mailto:lennart@karssen.org">lennart@karssen.org</a>>
            wrote:</div>
          <br class="Apple-interchange-newline">
          <blockquote type="cite">
            <div style="font-size: 12px; font-style: normal;
              font-variant: normal; font-weight: normal; letter-spacing:
              normal; line-height: normal; orphans: auto; text-align:
              start; text-indent: 0px; text-transform: none;
              white-space: normal; widows: auto; word-spacing: 0px;
              -webkit-text-stroke-width: 0px;">I had the same issues as
              Yurii reported:<br>
              - compilation warnings when installing<br>
              - the 'file already exists' warning: this was one of the
              reasons why I<br>
              renamed the filename in the 'out=' option in test.R<br>
              - I also got the warning about not-unique column names,
              but I assumed<br>
              that wasn't the issue we were trying to fix here, so I
              didn't mention it.<br>
              <br>
              Yurii, does the script finish when you change the output
              file name in<br>
              test.R?<br>
              <br>
              <br>
              <br>
              Lennart.<br>
              <br>
              On 11/25/2013 12:26 PM, Yury Aulchenko wrote:<br>
              <blockquote type="cite">Tried this on Mac OS X, see below<br>
                <br>
                On Nov 23, 2013, at 15:12 PM, Maksim Struchalin <<a
                  moz-do-not-send="true"
                  href="mailto:m.v.struchalin@mail.ru">m.v.struchalin@mail.ru</a><br>
                <<a moz-do-not-send="true"
                  href="mailto:m.v.struchalin@mail.ru">mailto:m.v.struchalin@mail.ru</a>>>
                wrote:<br>
                <br>
                <blockquote type="cite">Hi Lennart,<br>
                  <br>
                  <a moz-do-not-send="true"
                    href="https://app.box.com/s/iy41ug5qg4sbylul9oyn">https://app.box.com/s/iy41ug5qg4sbylul9oyn</a><br>
                  <br>
                  This is an example demonstrating how a GenABEL
                  function<br>
                  "impute2databel" calls a function "iteratorDA" from
                  DatABEL. Here,<br>
                  GenABEL is compiled without flv's and iterator's code
                  (I deleted it<br>
                  from src).<br>
                  <br>
                  Could you run the test?:<br>
                  0) Dowload the file test_GenABEL_iterator.tar.gz from<br>
                  <a class="moz-txt-link-freetext" href="https://app.box.com/s/iy41ug5qg4sbylul9oyn">https://app.box.com/s/iy41ug5qg4sbylul9oyn</a><br>
                  1) decompress test_GenABEL_iterator.tar.gz<br>
                  2) cd test_GenABEL_iterator<br>
                  3) R CMD INSTALL DatABEL_0.9-4.tar.gz<br>
                  4) R CMD INSTALL GenABEL_1.7-7.tar.gz<br>
                </blockquote>
                <br>
                getting many warnings at step (4)<br>
                <br>
                <blockquote type="cite">5) run test.R<br>
                </blockquote>
                <br>
                getting<span class="Apple-converted-space"> </span><br>
                <br>
                <blockquote type="cite">x <-
                  impute2databel(geno="TEST10x15.geno",
                  sample="impute.sample5",<br>
                </blockquote>
                out="TEST10x15_T.geno", makeprob=FALSE, old=TRUE)<br>
                Loading required package: DatABEL<br>
                DatABEL v.0.9-4 (March 12, 2013) loaded<br>
                <br>
                Options in effect:<span class="Apple-converted-space"> </span><br>
                --infile    = TEST10x15.geno<br>
                --outfile   = ./tmp333314<br>
                --skiprows  = OFF<br>
                --skipcols  = 5<br>
                --cnrow     = OFF<br>
                --rncol     = ON, using column 2 of 'TEST10x15.geno'<br>
                --transpose = ON<br>
                --Rmatrix   = OFF<br>
                --nanString = NA<br>
                Number of lines in source file is 10<br>
                Number of words in source file is 20<br>
                skiprows = 0<br>
                cnrow = 0<br>
                skipcols = 5<br>
                rncol = 2<br>
                Rmatrix = 0<br>
                numWords = 20<br>
                Creating file with numRows = 10<br>
                Creating file with numColumns = 15<br>
                text2fvf finished.<br>
                File 'TEST10x15_T.geno.dose' already exists.<br>
                ERROR in Rstuff:failed in ini_empty_FileMatrix_RError in
                !result :<br>
                invalid argument type<br>
                Calls: impute2databel -> apply2dfo ->
                make_empty_fvf<br>
                In addition: Warning message:<br>
                In uninames(.Object@data) :<br>
                 uninames: some column names are not unique; use<br>
                set_dimnames/get_dimnames for non-unique row/col names<br>
                Execution halted<br>
                <br>
                <br>
                <blockquote type="cite"><br>
                  It works on my Ubuntu. If it works on your Ubuntu, win
                  and mac, then<br>
                  we can delete from GenABEL the simlinks to flv and
                  databel.<br>
                  <br>
                  best,<br>
                  Максим<br>
                  <br>
                  <br>
                  On 22/11/2013 21:34, L.C. Karssen wrote:<br>
                  <blockquote type="cite">Hi максим,<br>
                    <br>
                    <br>
                    On 11/19/2013 03:17 PM, Maksim Struchalin wrote:<br>
                    <blockquote type="cite">Hi Lennart,<br>
                      <br>
                      I see you are improving your Russian :-).<br>
                    </blockquote>
                    Getting to know the Russian alphabet is step one
                    :-).<br>
                    <br>
                    <blockquote type="cite">I understand your arguments.
                      I think we can combine our two approaches.<br>
                      1) We make a so/dll from filevector and let it use
                      by<br>
                      ProbABEL/OmicABEL/Another_not_R_softABEL.<br>
                      2) For GenABEL and other R packages, we make a
                      DatABEL.<br>
                      <br>
                      The code of filevector is the same both for 1) and
                      2).<span class="Apple-converted-space"> </span><br>
                    </blockquote>
                    But that doesn't solve the problem of having
                    symlinks to the fvlib<br>
                    directory in our SVN tree... Which means that any
                    update to filevector<br>
                    can make the depending package (DatABEL) become
                    uncompilable.<br>
                    <br>
                    In the mean time I've set the first steps towards
                    'libfilevector' in<br>
                    SVN, see commits 1415 and 1416. This works (at least
                    for ProbABEL), but<br>
                    more polishing is needed.<br>
                    <br>
                    <br>
                    <blockquote type="cite">We only add<br>
                      preprocessor commands (#ifdef and so on) to
                      surround R specific code<br>
                      (ISNAN() and std::isnan). In this case, compiler
                      choose itself weather<br>
                      it buids the lib for R or for OS.<br>
                      <br>
                      If we will want to use only approach 1) for
                      GenABEL in the future, we<br>
                      can quickly swith to it later.<br>
                    </blockquote>
                    True, for now this will work.<br>
                    <br>
                    <br>
                    Best,<br>
                    <br>
                    Lennart.<br>
                    <br>
                    <blockquote type="cite">best,<br>
                      Maksim<br>
                      <br>
                      <br>
                      <br>
                      On 19/11/2013 16:10, L.C. Karssen wrote:<br>
                      <blockquote type="cite">Hi максим,<br>
                        <br>
                        (trying a Russian keyboard layout, no idea if
                        this works...).<br>
                        <br>
                        On 19-11-13 09:44, Maksim Struchalin wrote:<br>
                        <blockquote type="cite">It seems that your
                          solution is workable but I see little
                          difference with<br>
                          what it is now. Now the filevector code is
                          incorporated in each<br>
                          packages.<span class="Apple-converted-space"> </span><br>
                        </blockquote>
                        This is what I would like to change, indeed.
                        Code that is reused by so<br>
                        many packages should not be copied/symlinked
                        into the code tree of those<br>
                        packages. By symlinking it as we have now, there
                        is no proper way of<br>
                        specifying a version number of the filevector
                        code. Which, in turn means<br>
                        that if something changes in the filevector code
                        all other packages need<br>
                        to be changed immediately (just like what
                        happened with your latest<br>
                        change). If the filevector code have been a
                        proper library we could have<br>
                        simply said that ProbABEL still depends on the
                        old filevector version<br>
                        and take more time to make sure the two play
                        nice together.<br>
                        <br>
                        Moreover, with the filevector code in a separate
                        library the whole<br>
                        isnan() issue would not be a problem. We could
                        simply use std::isnan(),<br>
                        because CRAN wouldn't need to compile the
                        .so/.dll, so no need of ISNAN().<br>
                        When code is put in a library the internals
                        don't matter as long as the<br>
                        interface (function names + arguments) to the
                        outside doesn't change.<br>
                        <br>
                        <blockquote type="cite">You propose to follow
                          the same way but pack filelvector code<br>
                          in one file (dll or so) and distribute 9
                          packages form GenABEL with the<br>
                          same library.<br>
                        </blockquote>
                        Indeed. The problem with incorporating it all in
                        DatABEL is that non-R<br>
                        packages like ProbABEL and OmicABEL depend on
                        the stuff in the fvlib<br>
                        directory as well. Filevector is central to
                        (almost) all packages in the<br>
                        GenABEL suite, which is why I proposed to make a
                        library out of it. And,<br>
                        as noted above, this way packages can depend on
                        different version of the<br>
                        library.<br>
                        <br>
                        We can of course discuss whether we want to
                        distribute this .so/.dll as<br>
                        a separate (operating system) package or withing
                        the R packages. To me<br>
                        the first option is the 'correct' one, but I see
                        that this may impose on<br>
                        the user (except on Windows and maybe MacOS,
                        where the .so/.dll is<br>
                        included in the R package).<br>
                        <br>
                        <br>
                        <blockquote type="cite">Last time I proposed to
                          move filevector in DatABEL. All other packages<br>
                          (GenA and so on) will load DatAB in R and use
                          filevector fucntions from<br>
                          DatA. When DatABEL is loaded through
                          library(DatABEL), the file<br>
                          DatABEL.so is loaded as well.<br>
                        </blockquote>
                        I think this is what should be done with the
                        DAlib directory (another<br>
                        symlinked dir).<br>
                        <br>
                        <blockquote type="cite">Thus, you do not need to
                          ask users to<br>
                          install additional lib because it is in
                          DatABEL already. I think this is<br>
                          a workable approach that will allow us to
                          delete the filevector code (or<br>
                          filevector so/dll) from all the packages.<br>
                          <br>
                          <br>
                          This is some quote from the R manual how to
                          register functions to make<br>
                          it available from DatAB to GenAB:<br>
                          <br>
                          <br>
     _______________________________________________<br>
                          <br>
                          <br>
                               5.4 Registering native routines<br>
                          <br>
                          By ‘native’ routine, we mean an entry point in
                          compiled code.<br>
                          <br>
                          In calls to |.C|, |.Call|, |.Fortran| and
                          |.External|, R must locate the<br>
                          specified native routine by looking in the
                          appropriate shared<br>
                          object/DLL. By default, R uses the operating
                          system-specific dynamic<br>
                          loader to lookup the symbol in all loaded DLLs
                          and elsewhere.<br>
                          Alternatively, the author of the DLL can
                          explicitly register routines<br>
                          with R and use a single, platform-independent
                          mechanism for finding the<br>
                          routines in the DLL. One can use this
                          registration mechanism to provide<br>
                          additional information about a routine,
                          including the number and type of<br>
                          the arguments, and also make it available to R
                          programmers under a<br>
                          different name. In the future, registration
                          may be used to implement a<br>
                          form of “secure” or limited native access.<br>
                          <br>
_____________________________________________________<br>
                          <br>
                        </blockquote>
                        Hmm, I will have to think about this. This seems
                        to be about how R finds<br>
                        out in which DLL a function is found (and maybe
                        where the DLL is found<br>
                        in the filesystem). I think this is separate
                        from the point below, but<br>
                        I'm not sure.<br>
                        <br>
                        <blockquote type="cite">Your argument was from
                          "5.8 Linking to other packages: It is not in<br>
                          general possible to link a DLL in package
                          *packA* to a DLL provided by<br>
                          package *packB". *I do not quite understand
                          what they mean under 'link'.<br>
                          May be the mean link a library during
                          intsalltion?<br>
                        </blockquote>
                        Yes, as far as I understand, they mean linking
                        to a library during<br>
                        installation/compilation.<br>
                        <br>
                        <br>
                        Best,<br>
                        <br>
                        Lennart.<br>
                        <blockquote type="cite">best,<br>
                          Maksim<br>
                          <br>
                          <br>
                          <br>
                          On 19/11/2013 15:14, L.C. Karssen wrote:<br>
                          <blockquote type="cite">Hi Maksim,<br>
                            <br>
                            Good question... The idea is to generate a
                            .dll file for Windows, but<br>
                            I'm not sure what would be the best way to
                            distribute that. It would be<br>
                            interesting to see how other packages do
                            that. For example, the XML<br>
                            package depends on libxml2:<br>
                            <a moz-do-not-send="true"
                              href="http://cran.r-project.org/web/packages/XML/index.html">http://cran.r-project.org/web/packages/XML/index.html</a>
                            and the Rcurl<br>
                            package depends on libcurl:<br>
                            <a moz-do-not-send="true"
                              href="http://cran.r-project.org/web/packages/RCurl/index.html">http://cran.r-project.org/web/packages/RCurl/index.html</a><br>
                            <br>
                            In the XML package .zip file for Windows
                            there is a directory libs/x64<br>
                            and a directory libs/i386. Both contain
                            XML.dll, so I think that for<br>
                            Linux you simply specify a dependency on a
                            library, whereas for Windows<br>
                            the actual .dll is in the package (which is
                            quite logical because<br>
                            Windows lacks the package repositories that
                            most Linux distros have).<br>
                            It seems that for MacOS the .tgz file also
                            contains a lib directory with<br>
                            the .so file.<br>
                            <br>
                            <br>
                            Best regards,<br>
                            <br>
                            Lennart.<br>
                            <br>
                            On 19-11-13 08:56, Maksim Struchalin wrote:<br>
                            <blockquote type="cite">Hi Lennart,<br>
                              <br>
                              How the users under win will install such
                              a library?<br>
                              <br>
                              best,<br>
                              Maksim<br>
                              <br>
                              On 19/11/2013 14:46, L.C. Karssen wrote:<br>
                              <blockquote type="cite">Dear all,<br>
                                <br>
                                The Jenkins setup already shows its
                                value: After Maksim changed the call<br>
                                from std::isnan() to ISNAN() in fvlib's
                                CastUtils.cpp an automatic build<br>
                                of ProbABEL was triggered and it failed
                                (because ISNAN() is an R function).<br>
                                <br>
                                I guess this is one more reason to try
                                to convert fvlib into a real<br>
                                (shared) library.<br>
                                Does anyone have another workable
                                solution?<br>
                                <br>
                                <br>
                                Lennart.<br>
                                <br>
                                <br>
_______________________________________________<br>
                                genabel-devel mailing list<br>
<a class="moz-txt-link-abbreviated" href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                              </blockquote>
_______________________________________________<br>
                              genabel-devel mailing list<br>
                              <a class="moz-txt-link-abbreviated" href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                              <br>
                            </blockquote>
_______________________________________________<br>
                            genabel-devel mailing list<br>
                            <a class="moz-txt-link-abbreviated" href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                          </blockquote>
_______________________________________________<br>
                          genabel-devel mailing list<br>
                          <a moz-do-not-send="true"
                            href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                          <br>
                        </blockquote>
                        _______________________________________________<br>
                        genabel-devel mailing list<br>
                        <a moz-do-not-send="true"
                          href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                      </blockquote>
                      <br>
                      _______________________________________________<br>
                      genabel-devel mailing list<br>
                      <a moz-do-not-send="true"
                        href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                      <br>
                    </blockquote>
                    <br>
                    <br>
                    _______________________________________________<br>
                    genabel-devel mailing list<br>
                    <a moz-do-not-send="true"
                      href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                  </blockquote>
                  <br>
                  _______________________________________________<br>
                  genabel-devel mailing list<br>
                  <a moz-do-not-send="true"
                    href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
                  <<a moz-do-not-send="true"
                    href="mailto:genabel-devel@lists.r-forge.r-project.org">mailto:genabel-devel@lists.r-forge.r-project.org</a>><br>
                  <a moz-do-not-send="true"
href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                </blockquote>
                <br>
                <br>
                <br>
                _______________________________________________<br>
                genabel-devel mailing list<br>
                <a moz-do-not-send="true"
                  href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
                <a moz-do-not-send="true"
href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
                <br>
              </blockquote>
              <br>
              --<span class="Apple-converted-space"> </span><br>
              *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*<br>
              L.C. Karssen<br>
              Utrecht<br>
              The Netherlands<br>
              <br>
              <a moz-do-not-send="true"
                href="mailto:lennart@karssen.org">lennart@karssen.org</a><br>
              <a moz-do-not-send="true" href="http://blog.karssen.org/">http://blog.karssen.org</a><br>
              GPG key ID: A88F554A<br>
              -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-<br>
              <br>
              _______________________________________________<br>
              genabel-devel mailing list<br>
              <a moz-do-not-send="true"
                href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
              <a moz-do-not-send="true"
href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a></div>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
genabel-devel mailing list
<a class="moz-txt-link-abbreviated" href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a>
<a class="moz-txt-link-freetext" href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a></pre>
    </blockquote>
    <br>
  </body>
</html>