Survey Point File Formats and Field Order

Treat the layout and the reference as separate information

A point-file format describes how values are arranged. A coordinate reference describes what those values mean in space. PNEZD and PENZD establish a column order; they do not identify a projection, a datum, a zone or a height system. A table in metres can still be incompatible with another table in metres. Keep a short reference note with the file stating the horizontal system, axis convention, units, vertical reference, source and project revision.

Use the downloadable reference template as a checklist, not as proof that the information has been established. When a value is unknown, obtain it from the source rather than filling a plausible default. If the only available description is local coordinates, the origin, orientation and scale relationship may still need documentation. A format conversion cannot discover those missing facts or make unrelated point clouds compatible simply by putting their columns in the same order.

Read the letters before using a shortcut

P means point identifier, N Northing, E Easting, Z elevation and D description. PNEZD therefore puts Northing before Easting; PENZD puts Easting before Northing. NEZ omits the identifier and description. Other arrangements can be mapped explicitly if the tool supports the needed fields. A description can itself contain spaces, punctuation or a comma, so counting visible pieces of text is not a reliable substitute for parsing the specified delimiter.

For the same point, the PNEZD row 001,200,100,10,CONTROL corresponds to the PENZD row 001,100,200,10,CONTROL. Only the two horizontal columns exchange places. The point stays at Easting one hundred and Northing two hundred. Copying the row unchanged and merely selecting the other format would instead reinterpret its position. Verify this distinction with an asymmetric known control, because points with equal Easting and Northing hide an axis swap.

Choose a delimiter that survives descriptions

The shared text importer supports comma, semicolon and tab separation. A TXT extension does not say which of these separators is present. Inspect several rows, including one with a long description. For comma-delimited data, a description containing a comma needs quoting; a literal double quote inside a quoted field is represented by doubled quotes. Properly quoted multiline descriptions form one record, even though they occupy several physical text lines.

A whitespace-delimited vendor export needs a deliberate conversion to a supported delimiter. Blindly replacing every space with a comma can split a description into many unintended columns. Similarly, a decimal comma must not be confused with the field separator. Prepare a clear numeric representation and an explicit separator before import. In these tools, coordinate values use decimal points, and the text encoding is UTF-8; simply renaming a legacy encoded file does not convert its bytes.

State whether the first row is a heading

A header helps humans identify fields but does not override the mapping that you confirm. If the first line is actually a point, treating it as headings loses a record. If column names are treated as data, the checker will encounter nonnumeric coordinate strings. Compare the preview count with the source and inspect the first interpreted record. Blank lines and multiline quoted fields also mean that a source line number is not always the same as a sequential point count.

Map identifiers and optional fields only where they really exist. A three-column coordinate table should not be forced into a five-column template by inventing names or heights. Extra fields can carry useful context and should remain attached to their original rows when the chosen operation promises to preserve them. Check the selected export layout as well: an explicitly reduced three-column output cannot also preserve every metadata column in that point file.

Keep identifiers as identifiers

Leading zeros may distinguish a field naming convention even when a receiving application treats numeric names as equivalent. Preserve the original spelling of an identifier, and review conflict rules before renumbering or merging. The names 001 and 1 may collide in software that interprets both numerically. A change audit should retain the original source row so repeated names do not collapse two separate observations into one ambiguous entry.

Spreadsheet applications can interpret long digit strings, date-like names and formula-like text in surprising ways. Open the data with explicit column types where that matters. Inspection reports and instrument-ready point files serve different purposes: spreadsheet formula protection can alter text if it is added to a point identifier. Use the exporter intended for the destination and compare the actual file content, not just the way a spreadsheet chooses to display it.

Select an Excel worksheet deliberately

An XLSX workbook may contain a cover sheet, formulas, hidden intermediate tables and the actual deliverable. The shared importer lists worksheets and requires a deliberate choice; it does not assume the first sheet is the survey. Review the resulting column preview just as carefully as a text file. Stored numeric values, rather than rounded display labels, are used for numeric coordinates. Keep identifiers as text when their exact characters matter.

Formula, date and error cells are rejected instead of silently becoming coordinates or blank values. Resolve them in the source workbook and prepare ordinary values for the intended export. Cancelling a worksheet selection leaves you free to choose again, including the same file. The common point-file limit is ten MiB and one hundred thousand records, but specialized tools may apply smaller geometry limits. Their displayed limits take precedence over assumptions based on the file extension.

Match the output to the receiving application

Ask the recipient for column order, delimiter, header policy, identifier rules, units and required precision. The converter can reorganize columns and perform explicitly selected unit changes, but it does not establish a coordinate datum transformation. Test one known point after import into the destination. Also test an empty optional height, a leading-zero name and a description containing punctuation so that a successful first row does not hide a later mismatch.

CSV does not carry surface connectivity, DXF entity semantics or the folder and geometry model of KML. Use the relevant dedicated tool for those file types and keep its unsupported-object report. A LandXML point extraction is useful as a point list but is not a lossless copy of the original terrain surface. Deliver the appropriate geometry file when the next operation depends on triangles, ordered paths or entity properties that ordinary columns cannot express.

Questions and answers