Detect spanning in cell

[Vince]

I have a table like the one shown below. It's easy to see that 1992 spans
from Column 2 to column 4. When I copy it in (say) notepad, I get something
like
1992 2000

In other words, I lose two tabs that indicate a span. it should be
1992 2000

My question is, is it possiblet to write a VBA macro that would substitute a
tab (chr 9) for any spanning? If so, how would I do this? Converting table
to text does not help either. I wrote a program in perl that does all other
processing. All I need to be able to do is to indicate spans by tabs from
the word table. So, I should have a

\t 1992 \t \t 2000 \t \t.

I tried exploring the cells and other .tables properties but can't seem to
find the right syntax.

Thanks,

Vince


1992
2000


Number
Proportion of temporary

employment
Proportion of total employment
Number
Proportion of temporary employment
Proportion of total

employment

 

[Vince]

I found this. Thanks Jezebel.
The RowIndex and ColumnIndex properties provide the information you need,
although as you observe, it can get *very* complex, particularly if there
are cells merged vertically as well as horizontally, and if there are split
cells as well.


As a general approach --


1. Iterate the table to find the cell with the highest ColumnIndex value --
this determines the number of columns needed for the HTML table. (It might
be higher than the number of columns Word thinks is in the table, if there
are split cells.)


2. To check for horizontal merging, compare cell with cell.next -- if the
same rowindex then the number of horizontal cells merged into the first is
Cell2.ColumnIndex - Cell1.ColumnIndex; if not in the same row, then merge =
Column Count - ColumnIndex + 1


3. To Check for vertical merging, find the next cell with the same
ColumnIndex value, then compare the rowindex values.

Hello Everybody
I have set out to create a simple macro that converts Word tables to HTML
format.
I was albe to do it for tables with regular column and row numbers, but I
have come upon a stumbling block when trying to tacke merged cells ...
these are really a drag!
Looking for a hint on how determine how many columns (or rows) a merged
cell spans.
When saving a word doc as HTML tables are saved correctly, so I guess
there must be a way.
Is there anyone who has undertaken this arduous task and has come
victorious.
alfonso gonzales

--------------------------------------------------------------------------------

a.. Next message: Tom Ogilvy: "Re: VBA, Bookmarks, Word and Excel... help
please!"
b.. Previous message: Doug Robbins: "Re: VBA, Bookmarks, Word and Excel...
help please!"
c.. In reply to: alfonso gonzales: "determine a merged cell span column
value"
d.. Messages sorted by: [ date ] [ thread ]
We are proud to have Web Hosting and Rack Housing from 9 Net Avenue
Deutschland.

 

[Jezebel]

You can detect merged cells by iterating the cells of the table and checking
the ColIndex property in each case, comparing it with the ColIndex of the
previous and next cells. (if the table might have vertically merged cells,
you need to do likewise with the RowIndex property also).

You can also check the table's .Uniform property: if TRUE, there are no
merged or split cells in the table.

 

[Vince]

Thanks for your reply.

I am totally stuck here (coffee or loo break didn't help either). Here's my
code:


For i = 1 To ActiveDocument.Tables.Count

ActiveDocument.Tables(i).Select

If Not ActiveDocument.Tables(i).Uniform Then
For j = 1 To ActiveDocument.Tables(i).Rows.Count

For Each acell In ActiveDocument.Tables(i).Rows(j).Cells
acell.Select

MsgBox acell.Range.text & " " & acell.ColumnIndex & " " & "
" & acell.Next.ColumnIndex

Next
Next

End If

Next

The column index value seems to be the real value (as it appears in the
table) and the merged cells are ignored. What I mean is, supposing the row
has 7 cells:
CELLS: 1,1 1,2 - 1,3 1,4 - 1,5 1,5 - 1,6
COLUMNS: 1 2-3 4-5 5-6


I don't get "1 2 ; 2 4 ; 4 6; 6 1 " as my message box returns, I instead
get: "1 2; 2 3 ; 3 4 ; 4 1" meaning word pretends that the cells that were
merged aren't there. How can I detect the span then? Please help me figure
this out!

Thanks a lot!

Vince

 

[Jezebel]

You're quite right. Interesting. This has changed since Word 2000. It seems
that Word is now following the HTML table structure more closely. (Which is
no bad thing, all in all.)

As you say, if you start with a row containing seven cells and merge three
pairs, you have now have a row with four cells and nothing, it seems, to
indicate which cells were merged. Which is actually logical. Consider:

1. Create a table with several rows and seven columns.
2. In row one, merge cells 2-3, 4-5, 6-7
3. In row 2, merge cells 1-2, 3-4, 5-6
4. In row 3, merge cells 1-2-3-4
5. Now align the columns.

As a matter of logic, are those rows different? They all have four cells in
them, of equal width.

 

[Vince]

I see what you mean and would normally appreciate the way it works, but not
when I am trying to tag it. Anyway, I am on the way to building a solution
to this. (It's working with the tables I need to convert but I can imagine
many where it may fail). I am actually getting the smallest cell width (from
the row with the max number of columns) and dividing each cell's width with
it. Then I remove the tabs from the cells (if any) that I have split. After
splitting, I ensure that the new cells don't contain data and only contain a
tab. My Perl program then processes the whole thing! Kind of strange but
it's working so far!

For i = 1 To ActiveDocument.Tables.Count
Call DoCaptions("Processing table: " & i & " of " &
ActiveDocument.Tables.Count)
ActiveDocument.Tables(i).Select

If Not ActiveDocument.Tables(i).Uniform Then

Call DoCaptions("Calculating Maximum columns")

MaximumCols = ReturnMaxTableColumns(i, RowNumber)
MinColumnWidth = ReturnMinColumnWidth(i, RowNumber)

For j = 1 To ActiveDocument.Tables(i).Rows.Count
Call DoCaptions("Reading row : " & j & " of " &
ActiveDocument.Tables(i).Rows.Count & " in table " & i)

For Each aCell In ActiveDocument.Tables(i).Rows(j).Cells
aCell.Select

Counter = Int(aCell.Width / MinColumnWidth)
If Counter > 1 Then
'MsgBox "spanning: " & Counter & " rows"
Call DoCaptions("Killing span")
aCell.Split numcolumns:=Counter

'MsgBox aCell.Range.text
Suse = aCell.Range.text
Counter2 = Counter

While Counter2 > 1
Set bCell = aCell.Next
Counter2 = Counter2 - 1
bCell.Select
Suse = Suse & bCell.Range.text
'MsgBox bCell.Range.text
bCell.Range.text = vbTab
Wend
Suse = Replace(Suse, vbTab, "")
Suse = Replace(Suse, Chr(7), "")
Suse = Replace(Suse, Chr(10), "")
Suse = Replace(Suse, Chr(13), "")

aCell.Range.text = Suse

End If

Next
Next

End If

Next

 

[Klaus Linke]

Hi Vince,

For some other ideas, see http://www.word.mvps.org/faqs/macrosvba/GetRowColSpan.htm

But I would save as filtered HTML (or maybe XML in Word2003) and look at that.

BTW, the HTML format won't be fooled by Jezebel's table... it'll show no merged cells.
So probably Word figures out the rowspan and colspan cleverly when saving... It would just take some pretty sophisticated analysis of the table structure to do as well as the HTML export filters.

Usually you want to get the rowspan and colspan when you write some kind of export filter of your own. And you can save yourself a lot of work if you start from filtered HTML or WordML (XML) -- Although I myself don't do as I preach and use old macros inserting tags into the Word document, as I've used and perfected them since Word 6 ;-)

But one of these days, I'll throw them in the bin and do it properly... I swear.

Regards,
Klaus

 

[Vince]

Ha ha! I have a couple of macros too that deserved to be "binned". But, you
know, in a battle between business and idealism, business wins.

Thanks for the tip. I'll deviate from my current column-width analysis if it
proves to be useless in the long run. I hate to think about having to write
something again from scratch.

Thanks, again.

Vince

Hi Vince,

For some other ideas, see
http://www.word.mvps.org/faqs/macrosvba/GetRowColSpan.htm

But I would save as filtered HTML (or maybe XML in Word2003) and look at
that.

BTW, the HTML format won't be fooled by Jezebel's table... it'll show no
merged cells.
So probably Word figures out the rowspan and colspan cleverly when saving...
It would just take some pretty sophisticated analysis of the table structure
to do as well as the HTML export filters.

Usually you want to get the rowspan and colspan when you write some kind of
export filter of your own. And you can save yourself a lot of work if you
start from filtered HTML or WordML (XML) -- Although I myself don't do as I
preach and use old macros inserting tags into the Word document, as I've
used and perfected them since Word 6 ;-)

But one of these days, I'll throw them in the bin and do it properly... I
swear.

Regards,
Klaus

 

[Jezebel]

BTW, the HTML format won't be fooled by Jezebel's table... it'll show no
merged cells.

Indeed it won't. That was precisely my point. As far as HTML is concerned,
those rows are the same: four-column rows in a seven-column table. How they
got that way is irrelevant.