Issue 31866  —   Unifields: text files with frame definitions sometimes get corrupted.

Status:   Planned for resolution in 10.4.01

Solution available in patch(es):      10.3.02.001    9.7.05.014

Description:

 Summary:
 Uniface:
 Unifields: text files with frame definitions sometimes get corrupted.

 Environment:
 +Uniface Version  : 9.7.02 + G202 and higher / 10.3.01
 +Operating System : OS independent
 +Database         : DBMS independent
 +Browsers         : Browser independent

 Reproduction scenario
 1. A text file containing a frame defintion is loaded into an Unifield:
    lfileload "frame_test.txt", unifield , "UTF-8"
 2. A macro adding text is executed:
    macro "^ZOOMSome more text^SELECT^LAST_TEXT^SAVE^ACCEPT^USER_KEY^001"
 3. A text file is dumped via proc in the UserKey trigger:
    lfiledump $selblk, "framedumped.txt" , "UTF-8"

 Expected Result
 The same text file with text "Some more text" added

 Actual Result
 In the text fie some new WTY=xxx items in the frame definition,
 where xxx are random bytes.
 These bytes can be invalid UTF-8 byte sequences resulting in
 crashes or messages like
 0131 - Error on field XXX; one or more characters not found in character set.

Workaround:

 There is no known workaround for this problem.

Notes: