Bug 62740 - XSSFTable constructor automatically assigns invalid (non-unique) column IDs
Summary: XSSFTable constructor automatically assigns invalid (non-unique) column IDs
Status: RESOLVED FIXED
Alias: None
Product: POI
Classification: Unclassified
Component: XSSF (show other bugs)
Version: 4.0.0-FINAL
Hardware: All All
: P2 normal (vote)
Target Milestone: ---
Assignee: POI Developers List
URL:
Keywords:
: 62758 (view as bug list)
Depends on:
Blocks: 62758
  Show dependency tree
 
Reported: 2018-09-19 16:05 UTC by Andrew Frank Prusinowski
Modified: 2018-09-25 19:42 UTC (History)
1 user (show)


Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Andrew Frank Prusinowski 2018-09-19 16:05:57 UTC 
By default, XSSFTableColumn IDs have the non-unique value of '1'.  This is interpreted as an invalid table by Microsoft Office Excel.

The somewhat dated 'Create Table' example will reproduce this behavior: (https://svn.apache.org/repos/asf/poi/trunk/src/examples/src/org/apache/poi/xssf/usermodel/examples/CreateTable.java)

When specifying a table and auto-generating the columns via a cell range, XML similar to the following is produced (in 'table1.xml):

<?xml version="1.0" encoding="UTF-8"?>
<table id="1" ref="A1:C3" name="Test" displayName="Test_Table" xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"><tableColumns count="3"><tableColumn id="1" name="Column1"/><tableColumn id="1" name="Column2"/><tableColumn id="1" name="Column3"/></tableColumns><tableStyleInfo name="TableStyleMedium2" showRowStripes="true" showColumnStripes="true"/></table>

As can be seen, by default, all table columns are assigned a non-unique ID.

This behavior seems to stem from newly introduced logic (lines 285-293) in XSSFTable.java
Comment 1 Greg Woolsey 2018-09-19 17:34:06 UTC 
Yep, looks to me like the logic should start with 0 instead of 1, then add 1 after the loop.

I'm working on this now (in case other devs see it too).
Comment 2 Greg Woolsey 2018-09-19 17:34:06 UTC 
Yep, looks to me like the logic should start with 0 instead of 1, then add 1 after the loop.

I'm working on this now (in case other devs see it too).
Comment 3 Greg Woolsey 2018-09-19 17:43:46 UTC 
Fixed in r1841357.
Comment 4 Andrew Frank Prusinowski 2018-09-19 18:23:29 UTC 
Awesome; thanks for the good work :)
Comment 5 PJ Fanning 2018-09-25 19:42:57 UTC 
*** Bug 62758 has been marked as a duplicate of this bug. ***