openpyxl.preserve.crosspart

Targeted edits to non-worksheet parts under preserve mode.

Every part here is edited by byte splice against its ORIGINAL payload: appended entries land before the container's end tag, removed entries are cut at their scanned spans, and everything unmodeled (extLst, mc:AlternateContent, fileVersion, unknown attributes) passes through untouched. Rels are append-only with rId = max numeric existing + 1 ; [Content_Types].xml is edited by targeted append/remove, never regenerated (D12); calcChain removal cascades to its content-type override and workbook relationship (D13).

`CT_STYLESHEET_INDEX`

attributeCT_STYLESHEET_INDEX

= {tag: i for i, tag in (enumerate(CT_STYLESHEET_ORDER))}

`CT_STYLESHEET_ORDER`

attributeCT_STYLESHEET_ORDER

= ['numFmts', 'fonts', 'fills', 'borders', 'cellStyleXfs', 'cellXfs', 'cellStyles', 'dxfs', 'tableStyles', 'colors', 'extLst']

`CT_WORKBOOK_INDEX`

attributeCT_WORKBOOK_INDEX

= {tag: i for i, tag in (enumerate(CT_WORKBOOK_ORDER))}

`CT_WORKBOOK_ORDER`

attributeCT_WORKBOOK_ORDER

= ['fileVersion', 'fileSharing', 'workbookPr', 'workbookProtection', 'bookViews', 'sheets', 'functionGroups', 'externalReferences', 'definedNames', 'calcPr', 'oleSize', 'customWorkbookViews', 'pivotCaches', 'smartTagPr', 'smartTagTypes', 'webPublishing', 'fileRecoveryPr', 'webPublishObjects', 'extLst']

`WB_SPLICEABLE`

attributeWB_SPLICEABLE

= {'definedNames', 'calcPr', 'bookViews'}