User friendly workaround when trying to get collumns label with badly done table header

[pascalr0410]

This PR to implement a user friendly workaround when trying to get collumns label with badly done table header.

In place of the annoyings failleds assertions "Header cell can't be empty" and "Column labels must be unique" it implement the folowing bypass :

• Empty cell -> "#Empty"
• No_unique_label -> No_unique_label#2

Tests cases have been updated accordingly

[codecov]

Codecov Report

Merging #164 (db4e4f3) into master (d2f51a8) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #164   +/-   ##
=======================================
  Coverage   95.10%   95.11%           
=======================================
  Files          14       14           
  Lines        1879     1882    +3     
=======================================
+ Hits         1787     1790    +3     
  Misses         92       92           

Impacted Files	Coverage Δ
src/table.jl	94.93% <100.00%> (+0.06%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d2f51a8...db4e4f3. Read the comment docs.

[nilshg]

@felipenoris any chance to get this merged? The "column names must be unique" is the biggest usability issue I regularly encounter with XLSX, and adding a suffix to non-unique names seems an obvious (and widely used in other packages afaik) workaround.

[pascalr0410]

As me, you could use my personal repo at https://github.com/pascalr0410/XLSX.jl witch is a fork with patched collumn workarround by :

using Pkg
Pkg.rm("XLSX")
Pkg.add(url="https://github.com/pascalr0410/XLSX.jl")

But the best way would be to integrate my PR or a similar bypass.

[nilshg]

Thank you!