| Ticket UUID: | 5e6ae6e05ef86f233316cb0294a683afd3306a70 | |||
| Title: | Implement -strict correctly for cesu-8 | |||
| Type: | Bug | Version: | 8.7 | |
| Submitter: | jan.nijtmans | Created on: | 2023-02-16 11:11:25 | |
| Subsystem: | 11. Conversions from String | Assigned To: | jan.nijtmans | |
| Priority: | 5 Medium | Severity: | Important | |
| Status: | Closed | Last Modified: | 2023-02-16 21:32:23 | |
| Resolution: | Fixed | Closed By: | jan.nijtmans | |
| Closed on: | 2023-02-16 21:32:23 | |||
| Description: |
The cesu-8 encoding is slightly different from utf-8, which has a subtile effect on "-strict" Examples:
% encoding convertfrom -strict cesu-8 \xC0\x80
% encoding convertfrom -strict cesu-8 \xF1\x86\x83\x9C
Those 2 examples are illegal in cesu-8. \xC0\x80 is not illegal in modified cesu-8 (See TCL_ENCODING_MODIFIED flag, but this flag is not accessable at script level). The second example is a code point > 0xFFFF, which should be encoded as 2 surrogates in cesu-8. Expected:
% encoding convertfrom -strict cesu-8 \xC0\x80
unexpected byte sequence starting at index 0: '\xC0'
% encoding convertfrom -strict cesu-8 \xF1\x86\x83\x9C
unexpected byte sequence starting at index 0: '\xF1'
| |||
| User Comments: |
jan.nijtmans added on 2023-02-16 21:32:23:
Fixed [345299048d944cb3|here]. Closing. | |||
View Ticket