How to detect merged cells when reading tables

[sylvain-bougnoux]

Hello, I've created a simple .docx document (Word 2010) with a simple 2x2 table, the 1st col is merged as with:

+---+---+
|   | b |
+ a +---+
|   | c |
+---+---+

When reading this with python the cells (0,0) and (1,0) are different!
How can I detect that they are merged on the original document?
The text of both cell is indeed 'a'. But the 'pointer' are different.
Thanks
Python 2.7, docx 0.85

[sylvain-bougnoux]

I've found a workaround (though the doc does not stand).
With the above table in simple.docx

from docx import Document
doc=Document("simple.docx")
table=doc.tables[0]
c00=table.cell(0,0)
c10=table.cell(1,0)
c00==c01  # is false as reported
#however
c00._tc==c01._tc # is true

Regards

[oltish]

It has been a while, but someone else may need this information in future.

As mentioned in previous comment, docx.table._Cell object has property _tc which contains useful information about cell's span:

from docx import Document
doc=Document("simple.docx")
table=doc.tables[0]

c=table.cell(0,0)
print(c.text, c._tc.top, c._tc.bottom, c._tc.left, c._tc.right)

c=table.cell(0,1)
print(c.text, c._tc.top, c._tc.bottom, c._tc.left, c._tc.right)

c=table.cell(1,1)
print(c.text, c._tc.top, c._tc.bottom, c._tc.left, c._tc.right)

The output will be:

a 0 2 0 1
b 0 1 1 2
c 1 2 1 2

From this you will easily understand not only if the cell is merged or not, but also its shape and size.

[0-173]

I found that some docx files created by word have inconsistend grid_span attributes in a row.
When accessing a specific cell the ._cells property will not care about the rows defined by the xml but just count through the cells and assume that the cell count per row and all grid_span are correct.
In my case there were 7 columns in the table but one row contained only grid_span attribues of 1 + 1 + 2 + 2 = 6. After this row, all the succeeding rows are broken.
The workaround for detecting this has already been mentioned by @oltish.
I will provide a merge request as a fix for the _cells function.

[mrlnc]

The workaround from @0-173 works for me. Thanks! However, the new cell doesn't have any attributes like the "text" field.

python2
docx 0.8.10

[fusted2]

c00._tc==c01._tc # is true

Thank you for your idea. I create a list, and add cell_.tc, then check to skip the 2nd+ merged cells. This is my project:

https://github.com/fusted2/Copy-Word-tables-to-Excel