Re: Function as return value, how to document it (::usage)
- To: mathgroup at smc.vnet.net
- Subject: [mg97533] Re: Function as return value, how to document it (::usage)
- From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
- Date: Sat, 14 Mar 2009 18:15:57 -0500 (EST)
- References: <gpg1e9$cj4$1@smc.vnet.net>
Hi,
and what does
Information[Evaluate[myf]]
??
Regards
Jens
Karsten W. wrote:
> Hello,
>
> my function is a function builder, when it is called, it returns a
> function. Everything works fine, except for the help text.
> Here is a minimal example:
>
> funcBuilder[start_, factor_] := Module[
> {localfunc},
> localfunc[0] = start;
> localfunc[x_Integer] := localfunc[x - 1] + factor*x;
> localfunc[] := "empty";
> localfunc::usage = "my help text here.";
> localfunc
> ];
> myf = funcBuilder[1, 1];
> myf[]
> Table[myf[i], {i, 1, 5}]
> ?myf
>
> It displays "localfunc$7151" instead of "my help text here". Do you
> have a hint on how to improve this?
>
> To put this problem into a larger perspective, here is what I really
> want to do with this function builder. I need to read several Excel
> sheets and want a nice interface. I have some assumptions on the Excel
> sheets: data is rectangular, first row is column names,
> there is one column that has unique values (something like a primary
> key).
>
> Now I have defined a wrapper for the Import function as follows:
>
> ReadExcelSheet[file_, sheet_, pk_] :=
> Module[{firstRow, colNameRules, data, datafunc},
> data = Import[file, {"Sheets", sheet}];
> firstRow = First[data];
> data = Rest[data];
> colNameRules = MapThread[#1 -> #2 &, {firstRow, Table[i, {i, 1,
> Length[firstRow]}]}];
> datafunc[id_] := First[Select[data, Part[#, pk /. colNameRules] ==
> id &]];
> datafunc[id_, colName_String] := Part[datafunc[id], colName /.
> colNameRules];
> datafunc[id_, colNames_List] := Part[datafunc[id], colNames /.
> colNameRules];
> datafunc[] := colNameRules;
> datafunc[Null] := data;
> datafunc[Null, colName_String] := Map[ Part[#, colName /.
> colNameRules] &, data];
> datafunc[Null, colNames_List] := Map[ Part[#, colNames /.
> colNameRules] &, data];
> datafunc (* return value *)
> ];
>
> The function returns a function which allows data access like this
> (assuming we have a sheet with columns id and each month):
>
> ProductionData = ReadExcelSheet["myfile.xls", "mysheet", "id"]
> ProductionData["id1", "Jan"] (* returns one number *)
> ProductionData["id1", {"Jan", "Feb"}] (* returns two values *)
> ProductionData[Null, {"id", "Jan"}] (* returns id and january values
> for all plants *)
> ProductionData[Null] (*returns all values, but not the headers*)
> ProductionData[] (* returns transformation rules column names -> list
> index *)
>
> I hope that this approach does not have negative side effects ?!?
> Also, I would be interested in different approaches to this data
> access problem. And, I would like to add some information of the excel
> sheet (file and sheet name, timestamp, some usage information).
>
> Kind regards,
>
> Karsten.
>
>