[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #10 from Johannes Zarl-Zierl--- Git commit 61b306eaa1723ccd42fa926be1c2b2d7748f205e by Johannes Zarl-Zierl. Committed on 15/12/2016 at 01:27. Pushed by johanneszarl into branch 'master'. Reuse numerical tag ids. This patch makes KPA distinguish between tag order in the xml file and tag id. In most cases, tags will now get the same id when saving as they had when the db was loaded from disk. This allows us to reorder tags (as required by the SortLastUsed option of the ListSelect) and still have diffable xml files. M +1-1AnnotationDialog/ListSelect.cpp M +20 -5XMLDB/XMLCategory.cpp https://commits.kde.org/kphotoalbum/61b306eaa1723ccd42fa926be1c2b2d7748f205e -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #9 from Unknown--- Git commit 61fe41652be693d213ebbc3465bf970ce52a4f76 by Johannes Zarl-Zierl. Committed on 10/12/2016 at 23:37. Pushed by johanneszarl into branch 'master'. Fix the SortLastUsed functionality in ListSelect Unfortunately, sorting the ListSelect entries in the annotation dialog in a most-recently-used first order currently requires shuffling the tag list in the XMLCategory. That means that the numerical id's in the database invariably change to reflect the most-recently-used order... M +4-2XMLDB/XMLCategory.cpp https://commits.kde.org/kphotoalbum/61fe41652be693d213ebbc3465bf970ce52a4f76 -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 Johannes Zarl-Zierlchanged: What|Removed |Added Status|CONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #8 from Johannes Zarl-Zierl --- Fix was confirmed in pm to me... -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #7 from Johannes Zarl-Zierl--- (In reply to Johannes Zarl-Zierl from comment #6) > Git commit 69420290638781721128f2fd27b209829b73cd57 by Johannes Zarl-Zierl. > Committed on 09/12/2016 at 00:16. > Pushed by johanneszarl into branch 'master'. > > Add benchmarking code for deterministic saving of DB. > > On my system, deterministic saving costs about 34 msec with a database of > just below 25k images. > > Saving time, non-deterministic: 1473.2 msec > Saving time, deterministic: 1506.9 msec > > M +34 -0XMLDB/FileWriter.cpp > > https://commits.kde.org/kphotoalbum/69420290638781721128f2fd27b209829b73cd57 I forgot to mention in the commit message: the numbers were averages of 10 consecutive save operations... -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 Johannes Zarl-Zierlchanged: What|Removed |Added Latest Commit||https://commits.kde.org/kph ||otoalbum/728df5bb6073cb8d2d ||854daec1ecb26bfadd4f96 Resolution|--- |FIXED Status|CONFIRMED |RESOLVED Latest Commit|https://commits.kde.org/kph | |otoalbum/728df5bb6073cb8d2d | |854daec1ecb26bfadd4f96 | Status|RESOLVED|CONFIRMED Resolution|FIXED |--- --- Comment #5 from Johannes Zarl-Zierl --- Git commit 728df5bb6073cb8d2d854daec1ecb26bfadd4f96 by Johannes Zarl-Zierl. Committed on 09/12/2016 at 00:15. Pushed by johanneszarl into branch 'master'. Make saving the DB deterministic M +6-1XMLDB/FileWriter.cpp https://commits.kde.org/kphotoalbum/728df5bb6073cb8d2d854daec1ecb26bfadd4f96 --- Comment #6 from Johannes Zarl-Zierl --- Git commit 69420290638781721128f2fd27b209829b73cd57 by Johannes Zarl-Zierl. Committed on 09/12/2016 at 00:16. Pushed by johanneszarl into branch 'master'. Add benchmarking code for deterministic saving of DB. On my system, deterministic saving costs about 34 msec with a database of just below 25k images. Saving time, non-deterministic: 1473.2 msec Saving time, deterministic: 1506.9 msec M +34 -0XMLDB/FileWriter.cpp https://commits.kde.org/kphotoalbum/69420290638781721128f2fd27b209829b73cd57 -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 Johannes Zarl-Zierlchanged: What|Removed |Added Latest Commit||https://commits.kde.org/kph ||otoalbum/728df5bb6073cb8d2d ||854daec1ecb26bfadd4f96 Resolution|--- |FIXED Status|CONFIRMED |RESOLVED Latest Commit|https://commits.kde.org/kph | |otoalbum/728df5bb6073cb8d2d | |854daec1ecb26bfadd4f96 | Status|RESOLVED|CONFIRMED Resolution|FIXED |--- --- Comment #5 from Johannes Zarl-Zierl --- Git commit 728df5bb6073cb8d2d854daec1ecb26bfadd4f96 by Johannes Zarl-Zierl. Committed on 09/12/2016 at 00:15. Pushed by johanneszarl into branch 'master'. Make saving the DB deterministic M +6-1XMLDB/FileWriter.cpp https://commits.kde.org/kphotoalbum/728df5bb6073cb8d2d854daec1ecb26bfadd4f96 --- Comment #6 from Johannes Zarl-Zierl --- Git commit 69420290638781721128f2fd27b209829b73cd57 by Johannes Zarl-Zierl. Committed on 09/12/2016 at 00:16. Pushed by johanneszarl into branch 'master'. Add benchmarking code for deterministic saving of DB. On my system, deterministic saving costs about 34 msec with a database of just below 25k images. Saving time, non-deterministic: 1473.2 msec Saving time, deterministic: 1506.9 msec M +34 -0XMLDB/FileWriter.cpp https://commits.kde.org/kphotoalbum/69420290638781721128f2fd27b209829b73cd57 -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 Tobias Leupoldchanged: What|Removed |Added CC||tobias.leup...@gmx.de -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #4 from Johannes Zarl-Zierl--- Git commit d5f1ac71af2217f33df56b3c9739030ea2b60ae0 by Johannes Zarl-Zierl. Committed on 07/12/2016 at 01:16. Pushed by johanneszarl into branch 'master'. Make writing index.xml more deterministic. There were still a few QSet uses left that introduce non-determinism in file-writing. There is still some work to be done, but getting back diffable files seems viable. (Performance will also have to be checked, afterwards... M +9-2XMLDB/FileWriter.cpp M +2-3XMLDB/XMLCategory.cpp https://commits.kde.org/kphotoalbum/d5f1ac71af2217f33df56b3c9739030ea2b60ae0 -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #3 from Johannes Zarl-Zierl--- How is fb60b996 for a change? I've refrained from sorting the tags and member-groups in the way you describe, because this would mean that numerical tag id's change every time I add a tag (and thus almost every image entry would change as well). -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 --- Comment #2 from Johannes Zarl-Zierl--- Git commit fb60b9965e353e938507e1dd6c3c0ea636824235 by Johannes Zarl-Zierl. Committed on 07/12/2016 at 00:32. Pushed by johanneszarl into branch 'master'. Try harder to get minimal changes in the index.xml Scripts like kpa-backup require diff-able files if one wants human-readable change descriptions. This tweaks XMLCategory::addItem() so that it doesn't always change all numerical category ids, and sorts the blocklist before writing it to a file. M +4-1XMLDB/FileWriter.cpp M +2-2XMLDB/XMLCategory.cpp https://commits.kde.org/kphotoalbum/fb60b9965e353e938507e1dd6c3c0ea636824235 -- You are receiving this mail because: You are watching all bug changes.
[kphotoalbum] [Bug 370373] Please sort on saving index.xml
https://bugs.kde.org/show_bug.cgi?id=370373 Johannes Zarl-Zierlchanged: What|Removed |Added Status|UNCONFIRMED |CONFIRMED Ever confirmed|0 |1 CC||johan...@zarl-zierl.at --- Comment #1 from Johannes Zarl-Zierl --- The issue is known (and also a problem with the kpa-backup.sh script). I just didn't have time to fix it, yet. Expect a fix in version 5.1 (my apologies for not including it with 5.0)... -- You are receiving this mail because: You are watching all bug changes.