Rename substring in table

this is a table with field "Name"

i want to search if name exit in table

example:

name=Preset1

name exist in table ? yes

Then i rename "Preset1" in Preset1_1;

If Preset1_1 exist in table i rename in Preset1_2 ...if exist i rename in Preset1_3..(I add a number following the "_")

Answer 1

[Walter Roberson]

As a utility, see matlab.lang.makeUniqueStrings -- which already happens to append an underscore and a number. So if you were to make unique strings of name with respect to the existing entries in the Name field...

mask = YourTable.Name == string(name);
if any(mask)
    newname = matlab.lang.makeUniqueStrings(name, YourTable.Name);
    YourTable.Name(mask) = newname;
end

[Luca Re]

thank you walter but i've a problem:

xx.Name (xx is a table but xx.Name is a cell array)

ans =

3×1 cell array

{["Preset1"]}

{["Preset1"]}

{["Preset1"]}

pr =

"Preset1"

mask = xx.Name ==pr

Error using ==

Cell must be a cell array of character vectors.

[Walter Roberson]

mask = strcmp(YourTable.Name, name);
if any(mask)
    newname = matlab.lang.makeUniqueStrings(name, YourTable.Name);
    YourTable.Name(mask) = {newname};
end

[Luca Re]

the comparison works exclusively like this:

strcmp(string(Your table.Name), name);

thanks now it works

[Walter Roberson]

Ah I just noticed that your table contains a cell array of string() objects. That should have been avoided when constructing the table. You should either use a cell array of char vectors, or else a string() array.

If you are using something like readtable() to read from a xls or xlsx file, then by default text will be read in a a cell array of char vectors, but you can configure options to read it as a string() array instead. readtable() cannot be configured to read text entries as a cell array of string() array -- so for your table to have that column, the content would have had to have been computed and stored in the table. It should have been computed slightly differently.

[Luca Re]

the data comes from a structure..I don't have to read any data from .xls Anyway, thanks, that feature is special!

[Luca Re]

i use this code to assig to the table:

  pr=string(app.Preset_ListBox.Value);
  Preset=LoadPreset_Struct(pr);
  [~,Preset.Name,~]=fileparts(Preset.Name); 
dd=struct2table(Preset);  
  d1=convertvars(dd, dd.Properties.VariableNames(1:end), 'cell');
  if ~numel(app.Preset_UITable.Data)
      app.Preset_UITable.Data=d1;
  else
      app.Preset_UITable.Data(end+1,:)=d1;
  end    
  app.Preset_UITable.ColumnName = d1.Properties.VariableNames;
  
  

this is piece of my struct preset

[Walter Roberson]

struct2table converts both character vector fields and string fields, into variables that are string arrays. There is no reason to convertvars() them to cell: leave them a string arrays.

[Luca Re]

i convert it in cell because in Data there are string and number..

if I didn't convert to cell it would give me an error

  d1=struct2table(Preset);  
  %d1=convertvars(dd, dd.Properties.VariableNames(1:end), 'cell');
  if ~numel(app.Preset_UITable.Data)
      app.Preset_UITable.Data=d1;
  else
      app.Preset_UITable.Data(end+1,:)=d1;
  end    
  app.Preset_UITable.ColumnName = d1.Properties.VariableNames;
  
  
Error using  () 
Unable to perform assignment because value of type 'string' is not convertible to 'cell'.

[Walter Roberson]

  if ~numel(app.Preset_UITable.Data)

The recommended function for that test is isempty

[Luca Re]

correct execution!

[Walter Roberson]

i convert it in cell because in Data there are string and number..

I take it the error occurs in

      app.Preset_UITable.Data(end+1,:)=d1;

You would only get an error there if there is a field in the struct that can contain either string or numeric (possibly empty)

If that field that can potentially contain numeric information is the Name field, then you run into questions about what you are comparing, and about whether the input name can be numeric. If the input name is '1' and the table Name field contains numeric 1 then should that be a match? If the input name is '1.0' and the table Name field contains numeric 1 then should that be a match, since numerically 1.0 and 1 are the same value?

[Luca Re]

I made it a rule to call names with letters and possibly add final numbers at the end

[Walter Roberson]

But some of the struct 'Name' fields might contain numeric values? Which you are skipping over with an "I (personally) am not going to ever ask for those so it isn't going to be a problem" ? And when other people expect to be able to match the numeric entries in the Name fields ???

[Luca Re]

it could also be "Pres4et" Would there be any problems in this case?

i see Pre12set1 becomes Preset1 ok I'm fine with the result...at least I can see it by sight

[Walter Roberson]

I am still stuck on the question of whether LoadPreset_Struct(pr) can return a struct that has a numeric value in the Name field ?

If so then

  [~,Preset.Name,~]=fileparts(Preset.Name); 

would fail with an error message. It would only succeed if the Name field was character vector or cell array of character vector or string array.

[Luca Re]

Preset is a struct but Preset.Name is a string

whos Preset.Name

Preset =

struct with fields:

Name: "Preset3_1"

....