From: SourceForge.net <no...@so...> - 2006-02-09 14:53:41
|
Bugs item #1370444, was opened at 2005-11-30 18:01 Message generated for change (Comment added) made by byngl You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=384719&aid=1370444&group_id=25576 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Other Group: 5.9.4 Status: Open Resolution: None Priority: 5 Submitted By: lehi davis (lnxcwby) Assigned to: Martijn Verburg (karianna) Summary: [5.9.4] Export to TXT does not include CR/LF Initial Comment: The output from an "Export to TXT" of any character sheet (blank or populated) does not include appropriate CR/LF characters. The output is all on one line, and completely unreadable. The same character sheet in 5.9.3 produces the expected output. Tried using the template from 5.9.3 instead, but still no joy. ---------------------------------------------------------------------- >Comment By: Greg Bingleman (byngl) Date: 2006-02-09 09:53 Message: Logged In: YES user_id=328527 I think that this is more likely a side-effect of the removal of new lines after every token. (Oct 11, 2005) ---------------------------------------------------------------------- Comment By: Martijn Verburg (karianna) Date: 2006-02-08 10:59 Message: Logged In: YES user_id=252169 OK, I have a clue as to what is going on (and it's linked to the old '&' bug as well) - Looks like the encode writer is not called or an innocent CR/LF is given an encoding when it doesn't want one :-). This is going to take some careful debug to get right. - K ---------------------------------------------------------------------- Comment By: Eddy Anthony (eddyanthony) Date: 2006-02-08 10:03 Message: Logged In: YES user_id=886893 Opps, don't know how that happened, didn't mean to unassign you Kar :-) ---------------------------------------------------------------------- Comment By: Eddy Anthony (eddyanthony) Date: 2006-02-08 09:55 Message: Logged In: YES user_id=886893 I beleive the opposite is what has happened, it's not that it was removed from 5.8 but it was just recently added to the 5.9.x line. Try outputing to html and selecting the plain text OS, that should give you the same result. ---------------------------------------------------------------------- Comment By: Martijn Verburg (karianna) Date: 2006-02-08 09:54 Message: Logged In: YES user_id=252169 The TXT export is a new feature in the 5.9.x stream. I wrote the code originally but haven't touched it since 5.9.0 (hence me thinking it was a change in the Output Sheet since it was OK in 5.9.3 and not in 5.9.4). I'll look into it (but with my TM hat on I'm re-organising this whole project, so I'm not sure I'll get to it fast). - K ---------------------------------------------------------------------- Comment By: lehi davis (lnxcwby) Date: 2006-02-08 09:43 Message: Logged In: YES user_id=225722 I just fired up 5.8.1rc2 to test that, but the option for TXT export is not there. So much for that idea. Has that been removed from the tree completely, or just in the stable releases? ---------------------------------------------------------------------- Comment By: Eddy Anthony (eddyanthony) Date: 2006-02-07 16:42 Message: Logged In: YES user_id=886893 If the same OS works in 5.9.3 but (the sheet from the 5.9.3 releae) does not work in 5.9.4 then it is not the sheet but something in the code. I've noticed this problem as well. Justed tested this, 5.9.3 renders the sheet from 5.9.3 and 5.9.4 with no problems. 5.9.4 renders the sheet from 5.9.3 and 5.9.4 without returns, this is a code problem. ---------------------------------------------------------------------- Comment By: lehi davis (lnxcwby) Date: 2006-02-07 15:36 Message: Logged In: YES user_id=225722 :-) That makes MUCH more sense! I just tried copying outputsheets/d20/fantasy/text/csheet_plain.TXT from 5.9.3 and using it in 5.9.4 and get the same results... Garbled with no CR/LF. In fact, the diff between the two only shows a handful of changes: diff csheet_plain.TXT.593 csheet_plain.TXT.594 4a5 > |IIF(VAR.IF(var("UseAlternateDamage")==0;1;0):1)| 5a7,10 > |ELSE| > Vitality: |HP| > Wound Points: |ALTHP| > |ENDIF| ---------------------------------------------------------------------- Comment By: Andrew Wilson (nuance) Date: 2006-02-07 15:29 Message: Logged In: YES user_id=41667 OS in this case means Output Sheet, not Operating System. ---------------------------------------------------------------------- Comment By: lehi davis (lnxcwby) Date: 2006-02-07 13:15 Message: Logged In: YES user_id=225722 I see the behavior in both GNU/Linux (Fedora Core 4) and in Windows 2000. Both using J2SE/JRE 5.0 Update 6, so I don't think that it's an OS problem. I have not yet tried the new RC, but will do that soon. ---------------------------------------------------------------------- Comment By: Martijn Verburg (karianna) Date: 2006-02-07 11:46 Message: Logged In: YES user_id=252169 Hmm, if it was working in 5.9.3 that indicates it might be an OS problem - K ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=384719&aid=1370444&group_id=25576 |