WikiTraccs adds around 34 WT_ columns to the Site Pages library of every migration target site. After a completed migration you might want to remove columns you no longer need. This page describes which columns WikiPakk still needs, how it uses them, what happens when they are missing, and in which order to remove the rest.
Note
This page only matters for sites that were a migration target of WikiTraccs. If you use WikiPakk without a WikiTraccs migration, your Site Pages library has none of these columns, and there is nothing to remove.Only after the final cutover
WikiTraccs itself needs its columns and its content type for delta migrations, link fixing, page refinement, and repairs. Remove them only when you are sure that WikiTraccs will not run against the site again. The removal permanently deletes migration metadata.Cleanup script keep list
Older versions of the community cleanup script CleanSitePagesLibrary.ps1 keep only five columns and would deleteWT_In_CfCreated and WT_Out_CfTransformationTime. That deletion disables the WikiPakk page tree for the whole site. Before running any cleanup script, verify its keep list contains all seven columns from the table below.The Seven Columns WikiPakk Needs
WikiPakk reads the migrated page hierarchy from these Site Pages library columns:
| Internal Name | Type | How WikiPakk uses it | Effect of an empty value |
|---|---|---|---|
| WT_In_CfContentId | Single line of text | Identifies a migrated page and connects children to it; separates migrated from vanilla pages | Page loses its migrated identity and its children cannot link to it |
| WT_In_CfParentId | Single line of text | The parent edge of the migrated hierarchy; 0 marks a root page | Page is not placed correctly in the tree |
| WT_In_CfTitle | Single line of text | Search by the original Confluence title; also a source for the shown page label | Title search misses the page; the native Title or filename still supplies a label |
| WT_In_CfSiblingOrder | Number | Order of pages below the same parent | The migrated order is lost for that page |
| WT_In_CfSpaceKey | Single line of text | Admits migrated root pages; helps identify page copies via the filename | A migrated root page with an empty space key vanishes from the WikiPakk page tree, together with the tree branch below it |
| WT_In_CfCreated | Date and Time | Distinguishes original migrated pages from copies (compared with the SharePoint Created date) | Copy detection loses a signal and can misclassify the page |
| WT_Out_CfTransformationTime | Date and Time | Another original-versus-copy signal, including for pages without a FirstPublishedDate | Copy detection loses a signal and can misclassify the page |
Two consequences follow from how WikiPakk checks these columns:
One missing column disables the migrated hierarchy for the whole site. WikiPakk checks that all seven internal names exist in the Site Pages library. When even one is missing, WikiPakk ignores the hierarchy information in the Site Pages library for all pages of the site and shows the library’s pages as a flat list instead. Pages that carry the WikiTraccs content type do not even appear in that flat list. So migrated pages are gone from the WikiPakk page tree entirely. The pages themselves still exist in SharePoint and open via direct links, search, and the Site Pages library; it is the WikiPakk navigation that loses them. Only manual moves stored in the hidden hierarchy list still apply. Never delete a subset of the seven.
Values matter too. WikiPakk checks only that the seven columns exist, but the per-page behavior in the table’s last column depends on the stored values. Keep the values; do not blank them. The check compares exact internal names and does not check column types, so recreating a deleted column as a lookalike is risky: a wrong internal name fails the check, and a wrong type passes it and then breaks queries. Keeping the original columns is safer than recreating them.
For the reverse angle (which values to fill to build a hierarchy for script-created pages), see Scripting the Page Hierarchy.
The Columns That Can Be Removed
Every other WT_ column of the Site Pages library, meaning every WT_ column that is not in the seven-column table above, is not needed by WikiPakk (verified for WikiPakk 2.17.0) and can be removed.
Notes:
WT_In_CfModifiedis included in the WikiPakk diagnostics export when present. Removing it only reduces that support export by one value; navigation is unaffected.- “Not needed by WikiPakk” does not mean “not needed by anyone”. Your own list views, Power Automate flows, scripts, reports, or audit requirements may use these columns. Inventory such consumers before deleting.
The WikiTraccs Content Type
WikiTraccs assigns the content type Site Page (transformed by WikiTraccs) to every migrated page.
Who needs it:
- WikiTraccs needs it. Target site preparation checks for it, and stored-page handling, link fixing, and refinement classify pages by it. Keep it as long as WikiTraccs might run against the site again.
- WikiPakk does not need it in normal operation. With all seven columns present, WikiPakk identifies migrated pages by their columns and ignores content types. The content type name only plays a role in the degraded mode described above: when the seven-column check fails, pages with this exact content type name are excluded from the flat list that WikiPakk shows instead of the hierarchy.
| Seven columns | Page content type | Result |
|---|---|---|
| All present | Any | Migrated hierarchy works; the content type is irrelevant to WikiPakk |
| One or more missing | Site Page (transformed by WikiTraccs) | Migrated pages are gone from the WikiPakk page tree |
| One or more missing | Reassigned to the standard Site Page type | Pages appear in WikiPakk’s flat page list, but without their Confluence hierarchy |
Removing the content type (again, only after the final cutover): SharePoint refuses to delete a content type that items still use. Reassign every page to the standard Site Page list content type first, verify the seven column values survived the reassignment, then delete the list content type. If you keep the content type, remove its field links for the columns you delete first; SharePoint refuses to delete a column that a list content type references.
Safe Removal Sequence
- Confirm that no future WikiTraccs run (delta migration, link fixing, refinement, repair) is expected for the site.
- Inventory your own consumers of the columns: list views, flows, scripts, reports. Export column values if you need them as an audit record.
- Pilot the removal on one representative site and validate the page tree, breadcrumb, page order, copied pages, and multilingual pages.
- Optional, when removing the content type: reassign all pages to the standard Site Page list content type and verify the seven column values survived.
- When keeping the content type: remove its field links for the columns you delete.
- Verify the seven internal names from the table above still exist in the Site Pages library.
- Delete the other columns.
- Open the site in a fresh browser session and validate the page tree.