About TableWorkbench

TableWorkbench helps turn exported data into a table you can inspect, combine, compare, or pass to another application. Typical jobs include aligning supplier columns, checking two versions of a catalogue, separating a large export, or preparing contact records. Each tool focuses on a specific task so that you can choose the operation whose result you understand. The directory includes both general table operations and specialised workspaces for workbooks, nested data, historical matching, reconciliation, and import-file preparation.

The common thread is explicit decisions. A tool asks which columns matter, what a match means, and which output you want. Those choices can take a moment, but they make the result easier to explain later. A similar-looking value is not automatically the same entity. A blank is not automatically zero. A successful export is not automatically proof that the source was correct. The interface and help pages are designed to make those distinctions visible at the point where they affect your task.

General table tools read CSV, TSV, and pasted text. Dedicated workbook tools can read selected XLSX sheets and data ranges. Some specialised workspaces accept either delimited text or an XLSX range, while the nested-data converter works with JSON and JSONL. The calendar converter generates event files and the contact converter generates contact files from supported table inputs. Read the format labels and the individual tool page before choosing a source, because support is not identical across every workspace.

Workbook support concerns data rather than complete document reproduction. Styles, charts, macros, external links, and live spreadsheet calculation are not interchangeable with table values. Where a tool offers stored values, displayed text, or formula text, those choices answer different questions. A formula’s cached result may be old, and a workbook with unavailable results needs review. The tool does not silently calculate every spreadsheet formula or promise that a data export will preserve the appearance of the source workbook.

The general workflow starts with parsing confirmation, continues with explicit operation rules, and ends with result review and export. The original file on your device is preserved. Reports can explain modified cells, removed records, duplicate groups, unmatched keys, or unresolved values, depending on the operation. Use those reports to inspect more than the first few visible rows. A preview is helpful for orientation, while the full counts and exception reports describe the scope of the completed task.

For example, merging files stacks records and aligns their columns. Deduplication then addresses repeated keys using a policy you choose. Comparison describes differences against another table. These operations are separate because each makes a different decision about meaning and record count. A workflow can combine them, but keeping the steps understandable makes it easier to find the source of an unexpected result. It also helps you explain why a record was kept, changed, or excluded.

Delimited text is read as text before an operation asks for another interpretation. This helps preserve identifiers with leading zeros and long strings that should not become approximate numbers. Numeric and date operations use explicit conventions, including the relevant separators, formats, and missing-value rules. Changing the interface language does not translate your cell contents or select a different meaning for a date. A date that is ambiguous in the source remains a decision to resolve.

Likewise, similarity is a review aid rather than evidence of identity. Fuzzy matching presents candidates based on configured text comparison. Category standardization applies mappings you confirm. Duplicate consolidation exposes competing nonempty values instead of assuming which one is authoritative. These tools can reduce repetitive work, but they do not know whether two customers are the same person or which supplier’s value your organisation considers correct. You supply that context through rules and review.

The application reads and processes selected file contents in your browser. It does not have a table-upload service or a cloud project store for your imported records. Browser workers perform supported computation while the page manages the interface. Opening the site and loading its resources still require ordinary network requests, and the hosting provider can maintain access logs. Local processing describes where table work happens; it does not mean that visiting a website creates no network traffic.

Working sessions are held in memory. Export files you need before reloading, closing the tab, or clearing a workspace. Compatible tools in the general workbench can share an applied table, while specialised workspaces can have separate state and handoff controls. Do not assume a calendar or contact conversion is an automatically saved extension of every other workflow. The privacy page explains session lifetime, downloaded copies, rule files, and the distinction between application memory and files stored by your browser.

An on-screen result and a downloaded package can serve different purposes. A result table contains the selected output; a report explains exceptions or changes; a rules file describes configuration. Choose the appropriate scope and check its count. Export options can affect headers, separators, encoding markers, or formula-like values. These choices remain relevant when the next application opens the file, because the next application has its own import and type-detection behaviour.

Specialised outputs need an additional destination check. An event file can be structurally valid while displaying at an unexpected local time if the schedule’s time zone was wrong. A contact file can contain valid fields while the receiving address book handles duplicates differently than expected. A product CSV can pass supported checks while a store rejects an image or a shop-specific value. Try a small import and inspect it in the destination before adopting the entire result.

The site provides English and Simplified Chinese interfaces and help. Tool names and explanations are written around the task in each language: appending records, matching identifiers, splitting fields, or checking an export. The underlying data rules do not change with the interface. This lets a team discuss the same operation while using the language each person finds more comfortable. It also avoids treating an interface preference as a statement about a file’s date or decimal convention.

Help content includes practical examples, operational limits, and questions tied to the tool’s actual behaviour. It distinguishes a supported transformation from a feature that belongs in another application. If a label or explanation leads you to expect a different result, a small fictional example makes useful feedback. Include the tool, selected settings, expected output, and what you observed so that the issue can be understood without sharing a complete business file.

Choose one task and begin with the built-in example when the workspace provides one. Inspect its input and result before bringing in your own data. Next, try a representative subset containing the situations that matter to your work, such as duplicate keys, blank values, or quoted text. This is a way to learn the rule, not a substitute for checking the complete result. A sample cannot prove that every record in a larger source behaves identically.

Processing limits protect the browser from unexpectedly large work. Bytes, records, columns, cells, expanded archives, and output size can all matter. A compressed workbook can require much more memory after opening, and a join or reshaping operation can expand the result. Split work into coherent batches when needed and preserve a record of which inputs each batch contains. Keep original sources and download accepted results so that the browser session is never your only copy.