How does browser Excel comparison detect row and cell differences locally?
Dropping Original (Before) and Updated (After) files lets you compare Excel and CSV sheets inside a local Web Worker that parses worksheets, unifies header columns, matches records, and highlights changed cells inside browser memory. Because neither file touches a server, sensitive price books and ERP snapshots remain strictly on your device.
The worker inspects incoming spreadsheets using three safety-bounded stages built around the ECMA-376 OpenXML workbook specification:
- Multi-sheet workbook extraction: Files matching
.xlsxextensions or0x50 0x4BZIP signatures load every worksheet tab under180 MBunpacked XML,100 MBsingle-entry, and250,000row-per-sheet limits. - CSV and TSV decoding: Text exports strip leading UTF-8 BOM markers, decode UTF-8 with Windows-1252 fallback, and sniff comma, semicolon, tab, or pipe separators across
10,000characters and10logical rows. - Cross-format sheet selectors: Whenever either workbook contains multiple tabs, dropdown lists surface each sheet name and row count so you can compare an updated Excel tab straight against a legacy CSV dump.
Teams archiving individual worksheets in Git repositories before comparing quarterly revisions often extract standalone CSV tabs ready for version tracking alongside binary .xlsx snapshots.
Matching rows by key column versus row position with tolerance
Choosing how rows align decides whether your spreadsheet diff highlights meaningful record updates or floods your screen with line-shift noise. Clicking Load sample sheets loads catalog-before.csv and catalog-after.csv preconfigured with SKU key alignment and a 0.001 rounding tolerance to demonstrate both behaviors immediately.
Key-column alignment, duplicate keys, and unified header order
Matching by Key column indexes rows on a chosen column such as SKU, Order ID, or Email so inserted, deleted, or re-sorted rows never cause false edits below them. Duplicate keys automatically gain numbered disambiguation suffixes (#2, #3) and empty keys receive internal blank markers so repeated records never overwrite one another.
To compare schemas cleanly when columns move or get added, the unified header builder preserves all Before columns in their original order and appends any new After headers matched trimmed and case-insensitively. Delimited files loaded for key matching follow IETF RFC 4180 CSV specification quoting rules so embedded commas inside descriptions stay aligned cleanly.
Numeric tolerance, comma-number parsing, trim, and case rules
Each paired cell passes through Numeric tolerance evaluation first. Comma-formatted strings such as 1,250.50 normalize to floating-point numbers, treating two values as equal whenever their absolute difference stays at or below your tolerance (for example, 14.95 and 14.9504 match under 0.001).
When strings remain after numeric checks, Trim whitespace (enabled by default) collapses consecutive spaces and strips outer padding, while Ignore case lets you treat capitalization edits like Active versus active as unchanged.
| Comparison Setting | Default Behavior | Normalization Rule | Primary Purpose |
|---|---|---|---|
Key column matching |
Matches selected header name | Disambiguates duplicates with #2, #3 |
Prevents cascading diffs after insertions |
Row order matching |
Compares index i line by line |
Labels keys as Row 2, Row 3, etc. |
Compares keyless lists or fixed exports |
Numeric tolerance |
0 (exact match, configurable) |
Strips thousands commas before diff | Ignores ERP or pricing rounding noise |
Trim whitespace |
Enabled by default | Collapses \s+ and trims outer spaces |
Eliminates trailing-space false positives |
Ignore case |
Disabled (case-sensitive) | Compares lowercased normalized strings | Skips capitalization-only text edits |
For unstructured log files, SQL scripts, or plain JSON configuration snapshots rather than column tables, switch to a line-by-line plain text diff viewer cleanly.
Reviewing Added, Removed, and Modified rows in the diff grid
Once comparison completes, four KPI summary cards report Added rows (+N), Removed rows (-N), Modified rows (~N), and Unchanged rows side by side. Selecting Changed only, All rows, Added, Removed, or Modified filters the interactive table instantly.
KPI summary counters, row status filters, and 300-row viewer cap
The interactive Diff rows table highlights added records in soft emerald, removed rows in soft rose, and modified cells in amber displaying struck-through original values next to the updated value (before -> after). Up to 300 filtered rows render in the browser grid to keep scrolling responsive on large price lists without freezing browser layout painting.
Generated XML report cells follow W3C XML 1.0 specification escaping so ampersands, angle brackets, and quotes inside cell values never corrupt downloaded OpenXML worksheets. Unchanged cells retain plain text styling so stakeholder reviewers focus immediately on genuine pricing, quantity, or description edits.
Pre-comparison table consolidation and clean audit workflows
When revised inventory or invoice exports arrive split across regional batches, analysts frequently combine multiple CSV or Excel tables by key first so comparing before and after snapshots requires only one unified Excel diff pass. Grouping files beforehand ensures missing regional SKUs do not trigger fragmented partial diffs.
Exporting styled XLSX diff reports and formula-safe Diff CSV
Clicking Download Diff XLSX generates spreadsheet-diff-report.xlsx containing all compared rows regardless of the 300-row on-screen preview limit. Column widths auto-fit cleanly between 8 and 60 characters across both worksheets, and every workbook contains two OpenXML sheets built with custom <styleSheet> fills and bold Calibri 11 pt headers:
- Summary worksheet: Records Before and After filenames, selected worksheet names, total row counts, added, removed, modified, and unchanged KPI metrics, and exact changed-cell totals.
- Diff Details worksheet: Lists
Status,Row / Key, and every unified column using#F1F5F9gray headers,#DCFCE7light green added rows,#FEE2E2light red removed rows, and#FEF3C7amber modified cells formattingbefore -> afterdiffs.
Choosing Download Diff CSV outputs spreadsheet-diff-report.csv with a UTF-8 BOM and MITRE CWE-1236 formula injection guidance enabled, prefixing untrusted leading =, +, @, |, and non-numeric - cells with an apostrophe so desktop spreadsheet apps treat changed values as inert text rather than executable commands. Teams auditing updated meeting timetables exported from calendars can also pair our iCalendar schedule to spreadsheet converter before running a side-by-side schedule diff across quarterly shifts.