Finding wrong xml tag in a table

[tpk1982]

Hi,

It the second part of the thread XML Tage name  The sample file is attached in the previous mail.

Thanks Loic for your support on the thread. He provided me  script  with a logic of count of XML tag name. But in some cases it wont work.

So the revised thing is getting the header column xml tag name and check the other in below rows. If anything found then mark the words with red color.

Tried coding:

//removing unsed xml elements
var mains = function() {    var doc = app.properties.activeDocument,  
  xes, xe;  
  if (!doc) return;  
  xes = doc.xmlElements[0].evaluateXPathExpression(".//*");  
  while ( xe = xes.pop() ) {  
  xe.texts[0].isValid && !xe.texts[0].parentTextFrames.length && xe.remove();  
  }  
}  
var u;  
app.doScript("mains();",u,u,UndoModes.ENTIRE_SCRIPT); 
//main function
var main = function() {  
    myDoc=app.documents[0]
corrs= {  
"Item #":"sku",  
"Description":"description",   
"Brand":"brand",   
"Product Dimensions":"product_dimensions",   
"Quantity":"quantity",   
"List Price":"list_price",
"Pen Style":"pen_style",
"Nib Type":"nib_type",
"Nib Size":"nib_size",
"Ink Color":"ink_color",
"Color":"color"
};    
var sel = app.activeDocument.textFrames, tf, ps, ln,  n = 0, tableHeader, lines,errors = [], check;  
for(z=0;z<sel.length;z++){   
tf = sel;    
var headers = {count:0};
var xe = tf.properties.associatedXMLElement;    
if ( !xe ) return;    
ps = xe.xmlContent;    
tableHeading = xe.evaluateXPathExpression ( "//table_heading" );    
if ( !tableHeading.length ) return;    
ln = tableHeading[0].xmlContent.contents.split("\t");    
n = ln.length;    
while ( n-- ) {   
headers[corrs[ln]] = 0;    
headers.count+=1;    
}    
checkContents ( ps , headers ) ;
    }    
  }
function checkContents ( story , headers ) {    
var xes = story.associatedXMLElement.evaluateXPathExpression ( "//*[name(./..)='copy_frame']" ).reverse(),   
currIndex, currNumber, n = xes.length, i = 0, nXe, xEc, xEl, xeName, data = {}, lnCount = story.lines.length;  
var myarray=[];
var myarr=[];
for (k=0; k< story.lines.length; k++)
  {
           if ((story.lines.contents).search("Item #") != -1 )
           {
           lnCount=lnCount-1-k;
           }
           else
           continue;
    }
var row=0;
for (k=0; k< story.lines.length; k++)//row length
  {
      if ((story.lines.contents).search("Item #") != -1 )
           {
                   var mycolumnlnnth=story.lines.contents.split("\t");//column split
                   for(row=k+1;row<=lnCount;row++){
                      var values=story.lines[row].contents.split("\t");//row split
                        for(column=0;column<mycolumnlnnth.length;column++){
                        if(values[column].associatedXMLElements[0].markupTag.name == "table_subhead")  {continue};
                           if (values[column].markupTag.name != corrs[mycolumnlnnth[column]]) 
                           {                                                
                                alert("InvalidTag: "+mycolumnlnnth[row]);
                                mycolumnlnnth([row]).contents.fillColor="Red"
                           }
                        }
                    }
      }
    }
}  
var u;    
app.doScript ( "main()",u,u,UndoModes.ENTIRE_SCRIPT, "The Script" );   

Thanks,

K

[Loic.Aigon]

Ok,

what about this road then :

var main = function() {  
var sel = app.properties.selection, tf, ps, ln, headers = {count:0}, n 
if ( !sel || sel.length!=1 || !(sel[0] instanceof TextFrame) ) return
tf = sel[0];  
tf2 = tf.duplicate();
table = tf2.texts[0].convertToTable();
tf2.fit ( FitOptions.FRAME_TO_CONTENT );
check = checkTable(table);
tf2.remove();
alert(
( check.result )?
"all seems fine"
:
"Content is wrong\r"+check.errors.join("\r") 
);
}  
var u;  
var checkTable = function(table) {
var check = {result:false, errors:[]},
cols = table.columns.everyItem().getElements(), n = cols.length, nCo
while ( n-- ) {
nCol = cols;
colCheck = checkACol ( nCol );
if ( !colCheck.result ) check.errors.push ( "Missing values for "+col
}
check.result = check.errors.length==0;
return check;
}
var checkACol = function(col) {
var cells = col.cells.everyItem().getElements(), n = cells.length, nCe
ctrl.header = cells[0].contents;
i++;
while ( i<n ) {
nCell = cells;
nCell.contents=="" && ctrl.lacking.push ( (i+1) );
i++;
}
ctrl.result = (ctrl.lacking.length == 0 );
return ctrl;
}
app.doScript ( "main()",u,u,UndoModes.ENTIRE_SCRIPT, "The Script" );  

[tpk1982]

Hi Loic,

Thanks again for your reply in this thread. While running the above code throw undefined is not an object in line 48. Previously bracket missing in line 36, and ctrl is undefined, so i just added.

var main = function() {    
var sel = app.properties.selection, tf, ps, ln, headers = {count:0}, n   
if ( !sel || sel.length!=1 || !(sel[0] instanceof TextFrame) ) return  
tf = sel[0];    
tf2 = tf.duplicate();  
table = tf2.texts[0].convertToTable();  
tf2.fit ( FitOptions.FRAME_TO_CONTENT );  
check = checkTable(table);  
tf2.remove();  
alert(  
( check.result )?  
"all seems fine"  
:  
"Content is wrong\r"+check.errors.join("\r")   
);  
}    
var u;    
var checkTable = function(table) {  
var check = {result:false, errors:[]},  
cols = table.columns.everyItem().getElements(), n = cols.length, nCo  
while ( n-- ) {  
nCol = cols;  
colCheck = checkACol ( nCol );  
if ( !colCheck.result ) check.errors.push ( "Missing values for "+col ) 
}  
check.result = check.errors.length==0;  
return check;  
}  
var checkACol = function(col) {  
var cells = col.cells.everyItem().getElements(), n = cells.length, nCe, ctrl  
ctrl.header = cells[0].contents;  
i++;  
while ( i<n ) {  
nCell = cells;  
nCell.contents=="" && ctrl.lacking.push ( (i+1) );  
i++;  
}  
ctrl.result = (ctrl.lacking.length == 0 );  
return ctrl;  
}  
app.doScript ( "main()",u,u,UndoModes.ENTIRE_SCRIPT, "The Script" );    

Regards,

K

[Loic.Aigon]

While running the above code throw undefined

Guessing that the copy/paste failed because it was there in my editor. Anyway…

[tpk1982]

HI Loic,

Any suggestions with my OP code, I tried again with some changes but still not fixed my issue. Could you please suggest what I am doing wrong?

//removing unsed xml elements  
var mains = function() {    var doc = app.properties.activeDocument,    
  xes, xe;    
  if (!doc) return;    
  xes = doc.xmlElements[0].evaluateXPathExpression(".//*");    
  while ( xe = xes.pop() ) {    
  xe.texts[0].isValid && !xe.texts[0].parentTextFrames.length && xe.remove();    
  }    
}    
var u;    
app.doScript("mains();",u,u,UndoModes.ENTIRE_SCRIPT);   
//main function  
var main = function() {    
    myDoc=app.documents[0]  
corrs= {    
"Item #":"sku",    
"Description":"description",     
"Brand":"brand",     
"Product Dimensions":"product_dimensions",     
"Quantity":"quantity",     
"List Price":"list_price",  
"Pen Style":"pen_style",  
"Nib Type":"nib_type",  
"Nib Size":"nib_size",  
"Ink Color":"ink_color",  
"Color":"color"  
};      
var sel = app.activeDocument.textFrames, tf, ps, ln,  n = 0, tableHeader, lines,errors = [], check;    
for(z=0;z<sel.length;z++){     
tf = sel;      
var headers = {count:0};  
var xe = tf.properties.associatedXMLElement;      
if ( !xe ) return;      
ps = xe.xmlContent;      
tableHeading = xe.evaluateXPathExpression ( "//table_heading" );      
if ( !tableHeading.length ) return;      
ln = tableHeading[0].xmlContent.contents.split("\t");      
n = ln.length;      
while ( n-- ) {     
headers[corrs[ln]] = 0;      
headers.count+=1;      
}      
checkContents ( ps , headers ) ;  
    }      
  }  
function checkContents ( story , headers ) {      
var xes = story.associatedXMLElement.evaluateXPathExpression ( "//*[name(./..)='copy_frame']" ).reverse(),     
currIndex, currNumber, n = xes.length, i = 0, nXe, xEc, xEl, xeName, data = {}, lnCount = story.lines.length;    
var myarray=[];  
var myarr=[];  
for (k=0; k< story.lines.length; k++)  
  {  
           if ((story.lines.contents).search("Item #") != -1 )  
           {  
           lnCount=lnCount-1-k;  
           }  
           else  
           continue;  
    }  
var line_contents=[], column_name=[];
var values=[];
var row=0;
for (k=0; k< story.lines.length; k++)
  {
      if ((story.lines.contents).search("Item #") != -1 )
           {
                   line_contents=story.lines.contents;                   
                   column_name = line_contents.split("\t");
                   for(row=k+1;row<=lnCount;row++){
                      if(story.lines[row].words[0].associatedXMLElements[0].markupTag.name == "table_subhead")  {continue};
                      values=story.lines[row].contents.split("\t");
                      for(column=0;column<column_name.length;column++){
                            if (story.lines[row].words[column].associatedXMLElements[0].markupTag.name  != corrs[column_name[column]]) 
                           {                                                
                                alert("InvalidTag"+story.lines[row].words[column].contents);
                           }
                    }
                }
          }
    }
}    
var u;      
app.doScript ( "main()",u,u,UndoModes.ENTIRE_SCRIPT, "The Script" );  

[Loic.Aigon]

Hi

It seems that the indeces are wrong. Logging at your code returns this:

column:2

story:[object Story]

story.lines:[object Lines]

story.lines[row]:[object Line]

story.lines[row].words:[object Words]

story.lines[row].words[column]:[object Word]

story.lines[row].words[column].associatedXMLElements[0]:[object XMLElement]

story.lines[row].words[column].associatedXMLElements[0].markupTag.name:description

column_name:Item #,Description,Brand,Product Dimensions,Quantity,List Price

column_name[column]:Brand

corrs[column_name[column]]):brand

story.lines[row].words[column].contents:Fossball

ERROR

So on index 2, the tag name is description but the column_name[column] for that index return "Brand".

[tpk1982]

In my script of original post, I assume the string is not supported by associatedXMLElements[0].markupTag.name option. Please suggest how to convert those to object.