Read and Share CSV Files Without Changing Their Values

A CSV file, short for comma-separated values, is a text file that represents a table: commas separate fields into columns, and records represent rows. It carries field values rather than a complete spreadsheet workbook. To share one safely, export the intended table, check that its structure and values survived, and publish only data approved for public access.

Choose your next task to see how to handle the CSV without changing its values.

Commas And Quotes Define The Table

Here is a fictional dataset saved as monthly-visits.csv:

site_id,site_name,month,visits,note
00123,"Library, East",2026-09,1200,
00456,West Library,2026-09,0,Closed for renovation
00789,Central Library,2026-09,850,"Hosted ""Reading Day"""

The first record is a header row naming five columns. The next three records are observations. Each observation describes one library site in one month.

Stored Text	Meaning
"Library, East"	One field containing a comma
"Hosted ""Reading Day"""	The text Hosted "Reading Day"
The trailing comma after 1200	An empty final note field

Double quotes around a field keep a comma inside that value from creating another column. A literal double quote inside a quoted field is doubled. The first observation still has five fields, even though its final field is empty.

These conventions are documented in RFC 4180’s description of common CSV formatting. Fields containing line breaks also need quoting, so counting physical text lines is not always the same as counting records. Use a CSV-aware reader when checking record counts or field counts.

CSV Keeps Values, Not The Whole Workbook

CSV preserves field text, not a workbook’s full structure. Excel’s CSV export saves the text and values displayed in the active worksheet; formatting, graphics and other worksheet features are lost. Formula text can be exported when cells display formulas instead of results. Keep the original workbook if you need its calculations or presentation. Microsoft documents these export differences.

CSV also has no built-in mechanism to declare column data types or uniqueness rules, as the W3C’s CSV on the Web primer explains. The receiving application must interpret the text using its own settings or accompanying metadata.

For the sample above, document those meanings separately:

Column	Intended Meaning
site_id	Five-character text identifier; retain leading zeros
site_name	Site name as text
month	Reporting month in YYYY-MM form
visits	Whole-number count; zero is a confirmed count
note	Optional explanation; blank means no note supplied

This data dictionary distinguishes identifiers from quantities and explains what blanks mean. In this dataset, 00123 is a code, not a number to calculate with; 0 in visits is a confirmed count, not a missing value.

Import Settings Protect Identifiers And Dates

Use a text editor to inspect the actual commas, quotes and stored values. Use a spreadsheet application to browse the table—but check its import settings before trusting the display.

In Excel, Data → From Text/CSV provides an import preview where you can choose the delimiter. Opening a .csv directly instead uses Excel’s default data-format settings, which can reinterpret dates or identifiers with leading zeros. Microsoft’s import guidance describes both approaches.

Before loading, set identifier columns to text where the importer supports it. Then compare an exact value such as 00123 with the raw file. Adding CSV quotes alone does not declare a text data type. Use the leading-zero preservation checklist when codes must remain unchanged.

If everything appears in one column, check the delimiter. Some exports use semicolons rather than commas; Microsoft notes that Excel’s separator can vary with settings. Do not replace every comma manually: some belong inside quoted values.

Check The Export Against The Source Table

Prepare a dedicated publication table rather than exporting a presentation sheet full of titles, subtotals and explanatory rows. Use one clear header row and one consistent kind of observation.

In Excel, make the intended worksheet active, then use File → Save As and choose a CSV file type. Only the current worksheet is exported; other worksheets need separate files. Microsoft’s import and export guidance linked above describes this limitation.

Check the exported file against the intended source table—not just against how it looks when reopened:

Check	What Must Match
Structure	Every parsed record has the expected field count
Completeness	Observation count, excluding the header, plus the first and last observations
Values	Identifiers, dates, decimals, confirmed zeros and blanks
Text	Accented names and other non-ASCII characters

Use the UTF-8 export guide to prepare the publication file, and follow the CSV export validation workflow for a fuller comparison.

Privacy needs a separate review: remove sensitive information and confirm that the remaining data may be published publicly. A structurally correct CSV is not necessarily safe to release. If it contains untrusted text and recipients may open it in spreadsheet software, also check for formula injection.

Publish The Reviewed File, Then Inspect The Page

TablePage’s homepage describes CSV uploads that generate a public dataset page with a shareable link and filterable table. Upload only the reviewed publication file—not sensitive data or an unchecked working export.

Before distributing the link, inspect the resulting page. Check the same sample identifiers, counts and text you checked in the file, and verify numeric sorting where available.

Supply the source, reporting period, column definitions, missing-value conventions and reuse terms in accompanying documentation. The CSV carries the values; those definitions tell readers how to interpret and reuse them.