Fixed script write_LeoPyRef in in ecf7fe4c890bc. Now, LeoPyRef.leo has the 
same content no matter if it is saved from Leo or generated from 
write_LeoPyRef script.

On Saturday, December 18, 2021 at 7:24:37 PM UTC+1 Félix wrote:

> Thanks for looking into this. I've once signaled this problem last year, 
> and Edward fixed it - but it crept beck in at some point and indeed now 
> this problem is back again. 
>
> Thanks again also for finding out the way this could have happened ! 
> --
> Félix
>
> On Friday, December 17, 2021 at 10:55:34 AM UTC-5 vitalije wrote:
>
>> While working on python importer recently I've noticed strange behavior 
>> of the LeoPyRef.leo. If I just open this file using Leo and save it, I get 
>> a huge diff. I didn't have the time to investigate it further until now.
>>
>> By bisecting, I've found that since the revision b7d66e100 (which was at 
>> the 18th Oct 2020) every committed version of LeoPyRef.leo has the same 
>> issue. 
>>
>> The commit b7d66e100 added a button write-leoPyRef and I assume that 
>> since then this file was saved using this button. I can't see why this 
>> script produces different result than ordinary c.save, but it does. The 
>> tnodes in the LeoPyRef.leo produced by this script are not sorted according 
>> the tx attribute, like when the file is saved using normal c.save.
>>
>> This produced huge difference in this commit 8601 insertions and 680 
>> deletions, and yet only one small script was added.
>>
>> Unless tnodes are sorted, potentially every commit to this file can 
>> produce huge diffs.
>>
>> Vitalije
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/be572503-39b6-4874-9e3c-02ed261855d1n%40googlegroups.com.

Reply via email to