On this page
A Chinese class register, visitor list or workshop roster needs more than an alphabetical sort button. The pronunciation of a surname can differ from the same character in an ordinary word, and two people can have identical Chinese names. A usable result keeps each name attached to the right record while giving you a predictable order for finding it.
Start with a copy of the roster. Keep the Chinese name, a stable record identifier and the information that must move with that person on one row. Review uncertain pronunciations before sorting. The final check is not simply whether the first few names look alphabetical: every identifier should still accompany the same name and details as in the source.
Keep one person on each row
Put the Chinese name in the first column, followed by an internal identifier and any necessary group or attendance fields. In the ChinaClue sorter, columns are separated by tabs and records by line breaks. The first column supplies the pronunciation used for sorting; the complete row moves together. Copy the data rows from a spreadsheet without its heading row, because a heading such as 姓名 would otherwise be treated as another entry.
Do not sort a detached column of names and paste it back beside an unchanged identifier column. That creates a plausible-looking list with the wrong relationships. A group code, meal choice or attendance mark belongs to its record, not to its original screen position. If you need to remove information before sharing, remove the entire unnecessary column from a working copy while preserving the record identifier.
Names alone are poor identifiers. The fictional example below contains two people called 张安. Their identifiers distinguish them, and their group values differ. They must remain separate even though their name and generated sorting key match. Use your own permitted internal reference rather than a passport number or another sensitive identifier merely to make a list unique.
Fictional names and records for practice.
| Chinese name | Record ID | Group |
|---|---|---|
| 单宁 | CN-01 | A |
| 张安 | CN-02 | B |
| 欧阳明 | CN-03 | A |
| 曾琪 | CN-04 | C |
| 张安 | CN-05 | A |
| 吕文 | CN-06 | B |
Review surname readings
Select name mode for a list of people. It uses surname-aware pronunciation at the beginning of the name; phrase mode is for ordinary words and expressions. 单 can have a different reading as a surname, and 曾 also needs its name context. The example uses the surname readings shan for 单宁 and zeng for 曾琪. These are declared readings for fictional records, not a claim about every real person with those characters.
Ask the person or the roster owner when a reading is uncertain. Their confirmed name information takes precedence over an automatic suggestion. Correct the editable reading for that row, then inspect where the record moves. Keep a separate note of confirmed readings if you will rebuild the roster later. Editing the original input resets the tool's manual overrides, so finish the input rows before doing the final pronunciation review.
A compound surname is not two unrelated surnames. 欧阳 in 欧阳明 belongs together as the family name. The tool may display the reading as separate syllables for checking, but that display is not a ready-made spelling for a certificate or travel document. Chinese name spelling normally distinguishes the surname from the given name; document-specific spelling should be copied from the person's authoritative record, not invented from a sorting key.
Use a clear sorting convention
This tool uses an explicit A–Z key: it lowercases the reading, removes tone marks and spaces, and represents ü as v. That makes the result reproducible for the same input and corrected readings. It is not a claim to implement every dictionary's ordering rules, a stroke-count tiebreaker or the collation chosen by a particular spreadsheet. Use the same convention for all batches that will later be combined.
In this convention, 吕文 has the key lvwen, while a name read lu wen has luwen. The distinction between u and ü remains after tone marks are removed. A manual reading written with lǚ should therefore keep the same underlying key as lü or lv. Check these names explicitly when moving data from another system that writes the vowel differently; a spelling used on an official document need not be the same as this internal sorting representation.
When two keys are equal, the tool retains their original relative order, including when sorting descending. It does not choose between the two 张安 records by identifier or group. If your organisation requires a second sorting field, establish that order in the source first and check the equal-key group afterwards. A surname-based roster is useful for lookup, but it does not establish which person an ambiguous name refers to.
Separate duplicate rows from duplicate names
Leave duplicate removal off for the first pass. Compare the input and output record counts and inspect repeated names with their identifiers. An identical name is not evidence of an accidental copy. The two 张安 rows in the example should both remain because their identifiers and groups are different.
The duplicate-removal option compares complete trimmed rows, not just the name field. It can remove an exact repeated record, while keeping same-name rows with different details. Decide whether a repeated row is an error before enabling it: a repeated participant may legitimately have more than one event or session record. For that workflow, use a session identifier as part of the row and preserve both entries.
Check invisible formatting before treating two rows as the same. A fullwidth space, a Latin punctuation variant or a differently written identifier can hide an otherwise familiar entry. The fullwidth/halfwidth tool can help inspect form text, but do not automatically rewrite people's names or every field in a record. Keep an untouched source copy and apply only corrections whose meaning you have checked.
Export the complete records
Review all result pages rather than only the first screen. The sorter accepts up to 500 non-empty rows per batch, with a 30,000-character input limit and a 500-character limit per row. An unresolved Chinese character in the reading is flagged for review. It should prompt a pronunciation check before you regard the exported order as final, not an attempt to guess a missing reading from the record's position.
Copying or downloading TXT preserves the sorted rows and their tab-separated columns. This is the useful route when you want to paste the roster back into a spreadsheet. Select the destination's first cell and confirm that the name, identifier and group land in separate columns. If the spreadsheet imports everything into one column, choose a tab-separated import instead of changing the underlying name data.
The CSV export serves a different checking view: one field holds the original complete row and another holds its reviewed reading. Tabs inside that original-row field are still part of the field, rather than separate CSV columns. Use it to keep a pronunciation audit alongside the roster; do not assume it has the same column layout as the TXT output. Open Chinese text using a compatible Unicode encoding and check names before saving over another file.
Check the roster before sharing
For the fictional six-row example, the expected ascending order is 吕文, 欧阳明, 单宁, 曾琪, 张安, 张安. Confirm the identifiers in that same order: CN-06, CN-03, CN-01, CN-04, CN-02, CN-05. The two equal-key rows keep their original order. These checks catch both a sorting mistake and a name–identifier mismatch that a simple alphabetical glance could miss.
Keep the confirmed-reading copy with your working materials, and share only the fields the recipient needs. If names or membership change, rebuild from the current complete source and reapply the confirmed readings. Avoid appending a new sorted batch to an old one and assuming the combined result is still ordered. The paired roster template includes the fictional records, a place for reading decisions and a blank sheet for your own list.
Comments
Loading…
No comments yet.