D11604: kdirlistertest doesn't fail at random

jtamate updated this revision to Diff 30323. jtamate added a comment. QStringLiteral instead of QString. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11604?vs=30285=30323 REVISION DETAIL https://phabricator.kde.org/D11604 AFFECTED FILES

D11487: optimization of TextLineData::attribute

jtamate added a comment. > One question to that though: Why do you sort/lookup by `x.offset + x.length <= p`? Note how lower_bound returns the first iterator that is _not_ going to return true. Assuming there are neither overlaps nor unsorted entries. Lets call X the iterator returned

D11604: kdirlistertest doesn't fail at random

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY - Instead of testing the creation and deletion of one file, do it with 1000 files. - test the mime type

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#231522 , @mwolff wrote: > @jtamate looking at your screenshots, it represents closely what I see locally. Most notably, there are no red underlines in your screenshots which could arise due to spell checking.

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#231110 , @mwolff wrote: > Considering spell checking is involved - can you show a screenshot for how the file looks like for you? There shouldn't be a lot of spell checking going on, or so I hope...

D11487: optimization of TextLineData::attribute

jtamate added a comment. I'm sorry, the previous file lines are 10 times longer than expected. This one {F5761669 }can be managed by kate, not only by okteta. REPOSITORY R39 KTextEditor REVISION DETAIL https://phabricator.kde.org/D11487 To:

D11487: optimization of TextLineData::attribute

jtamate added a comment. Uploaded a similar file F5761614: fake.xml.gz It is created using http://interglacial.com/~sburke/pub/rtf2xml.html from a rtf with two images and some text. REPOSITORY R39 KTextEditor REVISION DETAIL

D10670: Reduce plasmashell frozen time

jtamate added a comment. In D10670#230938 , @hein wrote: > > This revision was not accepted when it landed; it landed in state Needs Review. > > Weird, David accepted it though? Yes. Shouldn't this be reported as a bug to

D10670: Reduce plasmashell frozen time

This revision was not accepted when it landed; it landed in state "Needs Review". This revision was automatically updated to reflect the committed changes. Closed by commit R242:33ddaaa23f6b: Reduce plasmashell frozen time (authored by jtamate). REPOSITORY R242 Plasma Framework (Library)

D11491: Don't calculate attribute() twice.

This revision was automatically updated to reflect the committed changes. Closed by commit R39:91999f6e9116: Dont calculate attribute() twice. (authored by jtamate). REPOSITORY R39 KTextEditor CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11491?vs=29927=30137 REVISION DETAIL

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#230895 , @mwolff wrote: > Looking at `TextLineData::addAttribute`, it doesn't seem to sort the data - how can you be sure that the attributes list really is sorted by `offset + length`? I think this should be

D11487: optimization of TextLineData::attribute

jtamate updated this revision to Diff 30134. jtamate added a comment. > Huh, just specify -DCMAKE_INSTALL_PREFIX to the same path where you install the KF5 devel packages to, as a simple fix. Thanks, that worked. > Are you really compiling in release mode while measuring this?

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#230755 , @mwolff wrote: > yes, definitely don't roll your own lower_bound - use the STL provided one. Are you really compiling in release mode while measuring this? Also, I can only repeat myself in saying

D11487: optimization of TextLineData::attribute

jtamate updated this revision to Diff 30117. jtamate edited the summary of this revision. jtamate added a comment. The problem with attribute() is that with a long line and word wrapping enabled, when the long line is shown, it is called as many times as the length of the line, in my

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#229906 , @mwolff wrote: > can you publish the test file and how you measured it? Then I can test it with perf too, to confirm the impact. I'm sorry, I can't. It contains too much private information to

D11487: optimization of TextLineData::attribute

jtamate added a comment. In D11487#229889 , @cullmann wrote: > Hmm, does it really get that much faster or is the stuff just moved to the subroutines? The sum of the subroutines time is lower than the routine time as shown in

D11487: optimization of TextLineData::attribute

jtamate updated this revision to Diff 29975. jtamate retitled this revision from "simple optimization of TextLineData::attribute" to "optimization of TextLineData::attribute". jtamate edited the summary of this revision. jtamate edited the test plan for this revision. jtamate added a comment.

D11491: Don't calculate attribute() twice.

jtamate created this revision. jtamate added reviewers: Frameworks, Kate. Restricted Application added projects: Kate, Frameworks. jtamate requested review of this revision. REVISION SUMMARY "kateTextLine" neither "i" shouldn't have changed since the first call to attribute two lines ago.

D11487: simple optimization of TextLineData::attribute

jtamate updated this revision to Diff 29925. jtamate edited the summary of this revision. jtamate edited the test plan for this revision. jtamate added a comment. Same results, but more concise syntax with the range-based for. REPOSITORY R39 KTextEditor CHANGES SINCE LAST UPDATE

D10742: get rid of the raw KFileItem pointers in KCoreDirListerCache

jtamate planned changes to this revision. jtamate added a comment. I've noticed this creates/exposes some crashes in other code/tests, for example in kdirmodeltest the second and later times the test is run. I'll check all the failing tests (given enough time). REPOSITORY R241 KIO

D11487: simple optimization of TextLineData::attribute

jtamate created this revision. jtamate added reviewers: Frameworks, Kate. Restricted Application added projects: Kate, Frameworks. jtamate requested review of this revision. REVISION SUMMARY Instead of getting the element to check 4 to 5 times each time, using the constant iterator, do it only

D11282: less expensive findByUrl in KCoreDirListerCache

jtamate updated this revision to Diff 29574. jtamate edited the summary of this revision. jtamate edited the test plan for this revision. jtamate added a comment. Depends on https://phabricator.kde.org/D10742, including the diff. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE

D10742: get rid of the raw KFileItem pointers in KCoreDirListerCache

jtamate updated this revision to Diff 29572. jtamate edited the summary of this revision. jtamate edited the test plan for this revision. jtamate added a comment. Pass all the unittests, without modifications, finally. Use as much as possible the recently added move semantics in KFileItem.

D11282: less expensive findByUrl in KCoreDirListerCache

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY From O(n) in the worst case to at most Log2(n) + O(1) comparisons. Unfortunately, fetching the files in a

D1231: Add Remote Access interface to KWayland

jtamate added a comment. As I don't see anything related to security in this patch, I have two questions. Could anyone with access to server:port manage the server wayland sessions or just create a new session? The access control should be done in the firewall? REPOSITORY R127

D10669: Reduce plasmashell frozen time

This revision was automatically updated to reflect the committed changes. Closed by commit R278:ff6c795dcda7: Reduce plasmashell frozen time (authored by jtamate). CHANGED PRIOR TO COMMIT https://phabricator.kde.org/D10669?vs=27575=29346#toc REPOSITORY R278 KWindowSystem CHANGES SINCE LAST

D11260: avoid asking for an empty protocol

This revision was automatically updated to reflect the committed changes. Closed by commit R313:a3364a87dda4: avoid asking for an empty protocol (authored by jtamate). REPOSITORY R313 KHtml CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11260?vs=29316=29343 REVISION DETAIL

D11260: avoid asking for an empty protocol

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY Another case of empty protocol for findProtocol. https://phabricator.kde.org/D11013 In this case, just

D9829: Don't create 2 msgbox (one of them crashes) when can't change rights

jtamate abandoned this revision. jtamate added a comment. With the changes in https://phabricator.kde.org/D10376 this patch is not needed. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D9829 To: jtamate, #frameworks, anthonyfieroni, dfaure Cc: ngraham, michaelh

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate abandoned this revision. jtamate added a comment. With the changes in https://phabricator.kde.org/D10376 this patch is not needed. REPOSITORY R244 KCoreAddons REVISION DETAIL https://phabricator.kde.org/D9862 To: jtamate, #frameworks, dfaure Cc: michaelh

D11149: Create a constructor for KLocalizedStringPrivate

This revision was automatically updated to reflect the committed changes. Closed by commit R249:a02e99069e15: Create a constructor for KLocalizedStringPrivate (authored by jtamate). REPOSITORY R249 KI18n CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11149?vs=28990=29029 REVISION

D11132: Avoid an asan runtime error

This revision was automatically updated to reflect the committed changes. Closed by commit R302:af5397635c4f: Avoid an asan runtime error (authored by jtamate). REPOSITORY R302 KIconThemes CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11132?vs=29002=29026 REVISION DETAIL

D11132: Avoid an asan runtime error

jtamate updated this revision to Diff 29002. jtamate added a comment. Added the link. REPOSITORY R302 KIconThemes CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11132?vs=28991=29002 REVISION DETAIL https://phabricator.kde.org/D11132 AFFECTED FILES src/kiconeffect.cpp To:

D11132: Avoid an asan runtime error

jtamate updated this revision to Diff 28991. jtamate added a comment. with spaces after : and , REPOSITORY R302 KIconThemes CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11132?vs=28946=28991 REVISION DETAIL https://phabricator.kde.org/D11132 AFFECTED FILES

D11149: Create a constructor for KLocalizedStringPrivate

jtamate updated this revision to Diff 28990. jtamate added a comment. With spaces after : and , REPOSITORY R249 KI18n CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11149?vs=28988=28990 REVISION DETAIL https://phabricator.kde.org/D11149 AFFECTED FILES

D11149: Create a constructor for KLocalizedStringPrivate

jtamate created this revision. jtamate added a reviewer: Frameworks. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY The constructor avoids the following asan runtime errors kdesrc/frameworks/ki18n/src/klocalizedstring.cpp:179:7:

D11132: Avoid an asan runtime error

jtamate added inline comments. INLINE COMMENTS > apol wrote in kiconeffect.cpp:44 > To what does this initialize the array? effect: 0 value: 0 color: QColor(Invalid) trans: false key: "" color2: QColor(Invalid) REPOSITORY R302 KIconThemes REVISION DETAIL

D11132: Avoid an asan runtime error

jtamate updated this revision to Diff 28946. jtamate added a comment. Cleaner initialization, thanks anthonyfieroni. REPOSITORY R302 KIconThemes CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11132?vs=28943=28946 REVISION DETAIL https://phabricator.kde.org/D11132 AFFECTED

D11132: Avoid an asan runtime error

jtamate created this revision. jtamate added a reviewer: Frameworks. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY Initialize the arrays to the maximum size of the enums and use c++11 initialization to ensure that all values are

D11067: Fix more cases of incorrect parameter to findProtocol

This revision was automatically updated to reflect the committed changes. Closed by commit R241:c25b6745366f: Fix more cases of incorrect parameter to findProtocol (authored by jtamate). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11067?vs=28801=28855

D10989: Check for nullptr in indexForNode

jtamate added a comment. In D10989#219594 , @mpyne wrote: > I think valgrind has some flags you can use to try to generate a backtrace for where a freed block of memory was freed from, which might be useful for debugging. Thanks. The

D11067: Fix more cases of incorrect parameter to findProtocol

jtamate updated this revision to Diff 28801. jtamate retitled this revision from "Fix 3 more cases of incorrect parameter to findProtocol" to "Fix more cases of incorrect parameter to findProtocol". jtamate added a comment. David's comment done. REPOSITORY R241 KIO CHANGES SINCE LAST

D11089: KUrlCompletion: early return if the URL is invalid like ":/"

jtamate accepted this revision. jtamate added a comment. This revision is now accepted and ready to land. I'll change https://phabricator.kde.org/D11067. REPOSITORY R241 KIO BRANCH master REVISION DETAIL https://phabricator.kde.org/D11089 To: dfaure, jtamate Cc: #frameworks, michaelh

D11067: Fix 3 more cases of incorrect parameter to findProtocol

jtamate added inline comments. INLINE COMMENTS > dfaure wrote in kurlcompletion.cpp:626 > I'm not 100% sure about this one because kurl() is only supposed to be called > if url.isURL(). > > Also, I tried to reproduce this assert with a unittest for kcompletion, but > it works here !? > >

D11067: Fix 3 more cases of incorrect parameter to findProtocol

jtamate added a comment. The backtraces: deleting characters from the beginning #10 0x7fdc893f249a in KProtocolInfoFactory::findProtocol (this=0x7fdc896b9720 <(anonymous namespace)::Q_QGS_kProtocolInfoFactoryInstance::innerFunction()::holder>, protocol=...) at

D11067: Fix 3 more cases of incorrect parameter to findProtocol

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY deleting characters from the beginning in a url in dolphin trying to use this url: ~:/ REPOSITORY R241

D11044: Don't try to find an icon for an empty url

This revision was automatically updated to reflect the committed changes. Closed by commit R241:8ba93b82e451: Dont try to find an icon for an empty url (authored by jtamate). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11044?vs=28690=28739 REVISION DETAIL

D10989: Check for nullptr in indexForNode

jtamate added a comment. > I tried testing > > kdialog --getexistingdirectory smb:// > > but this gives an error here "Unable to find any workgroups in your local network." even with samba started locally. Don't know why. I get always a crash trying kdialog

D11044: Don't try to find an icon for an empty url

jtamate updated this revision to Diff 28690. jtamate added a comment. Checking for empty url at the top of the method. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D11044?vs=28680=28690 REVISION DETAIL https://phabricator.kde.org/D11044 AFFECTED FILES

D11044: Don't try to find an icon for an empty url

jtamate added a comment. In D11044#218896 , @dfaure wrote: > Wait, who's calling KIO::iconNameForUrl() with an empty URL? This should be an assert or an early-return at the top of that method, not nested into one specific if(). Even the very

D11013: Detect incorrect paramenter in findProtocol

jtamate added a dependent revision: D11044: Don't try to find an icon for an empty url. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D11013 To: jtamate, #frameworks, dfaure Cc: michaelh

D11044: Don't try to find an icon for an empty url

jtamate added a dependency: D11013: Detect incorrect paramenter in findProtocol. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D11044 To: jtamate, #frameworks, dfaure Cc: michaelh

D11044: Don't try to find an icon for an empty url

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY The message kf5.kio.core: Refilling KProtocolInfoFactory cache in the hope to find "" was produced when

D11013: Detect incorrect paramenter in findProtocol

jtamate updated this revision to Diff 28679. jtamate retitled this revision from "Remove trailing data in the protocol parameter in findProtocol" to "Detect incorrect paramenter in findProtocol". jtamate edited the summary of this revision. jtamate edited the test plan for this revision. jtamate

D10742: get rid of the raw KFileItem pointers in KCoreDirListerCache

jtamate added a comment. In D10742#218222 , @markg wrote: > In D10742#218171 , @dfaure wrote: > > > Can someone explain to me how switching from pointers to values is making anything faster, or is a

D10989: Check for nullptr in indexForNode

jtamate added a dependency: D11013: Remove trailing data in the protocol parameter in findProtocol. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D10989 To: jtamate, #frameworks, dfaure Cc: michaelh

D11013: Remove trailing data in the protocol parameter in findProtocol

jtamate added a dependent revision: D10989: Check for nullptr in indexForNode. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D11013 To: jtamate, #frameworks, dfaure Cc: michaelh

D11013: Remove trailing data in the protocol parameter in findProtocol

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY The paramter can receive things like: - smb:/// - zip:/home/user/Download/.../Encoder.php

D10989: Check for nullptr in indexForNode

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY CCBUG: 390288 Checking for nullptr will avoid the crash and return an empty QModelIndex. TEST PLAN

D10857: Change qSort to std::sort in simplifiedUrlList

jtamate added a comment. I've found the problem. The problem is qSort itself. Chaning qSort to qStableSort I've got normal times in the test case, select 50.000 files and pressing Shift+Supr. Also in the small test, the times, running under callgrind are: qSort: between 33148 and

D10857: Change qSort to std::sort in simplifiedUrlList

jtamate added a comment. In D10857#214607 , @dfaure wrote: > I'm not opposed to the idea, but measuring CPU usage is a rather misleading indicator. What if it takes 10 times longer, because it's progressing much more slowly? ;) > > Please

D10537: Convert from old connect syntax to new one

This revision was automatically updated to reflect the committed changes. Closed by commit R241:ec296c0ff511: Convert from old connect syntax to new one (authored by jtamate). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10537?vs=28102=28136 REVISION DETAIL

D10537: Convert from old connect syntax to new one

jtamate updated this revision to Diff 28102. jtamate added a comment. Fixed David comments. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10537?vs=27238=28102 REVISION DETAIL https://phabricator.kde.org/D10537 AFFECTED FILES src/core/listjob.cpp

D10857: Change qSort to std::sort in simplifiedUrlList

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY qSort is depecreated in Qt5. But qSort is also quite slow compared to std::sort. There are 11 more

D10124: Faster simplejob start

jtamate added a comment. In D10124#213918 , @dfaure wrote: > This commit leads to > > 20:29:06.184 okteta(12932) QObject::connect|?libKF5KIOCore.so.5? QObject::connect: No such slot KIO::ListJob::slotTotalSize(KIO::filesize_t) in

D10742: get rid of the raw KFileItem pointers in KCoreDirListerCache

jtamate updated this revision to Diff 27909. jtamate edited the test plan for this revision. jtamate added a comment. The first patch had 12 unit test failing. This one has 2 1/2 unit test failing. I've changed findByUrl, instead of removing the item from the list, tell the item to

D10702: Always use a job to delete files to avoid freezing process waiting on IO

jtamate requested changes to this revision. jtamate added a comment. This revision now requires changes to proceed. My guess is that the problem is not in the direct delete of local files. In fact, unlink() in most of the modern filesystems is not affected by the size of the file and is

D10742: get rid of the raw KFileItem pointers in KCoreDirListerCache

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY This summary will change when the doubts are resolved (if they can be resolved!). Implement the first

D10670: Reduce plasmashell frozen time

jtamate updated this revision to Diff 27672. jtamate edited the summary of this revision. jtamate added a comment. Updated the summary and added a comment REPOSITORY R242 Plasma Framework (Library) CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10670?vs=27576=27672 REVISION

D10670: Reduce plasmashell frozen time

jtamate added a comment. In D10670#209823 , @mwolff wrote: > ok, looking at the reasoning in the other commit: > > - you need to extend the commit message here Will be done. > - you need to provide a comment in the code the

D10627: Reduce plasmashell frozen time to almost nothing

This revision was automatically updated to reflect the committed changes. Closed by commit R120:e1fe26f1fc78: Reduce plasmashell frozen time to almost nothing (authored by jtamate). REPOSITORY R120 Plasma Workspace CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10627?vs=27453=27577

D10670: Reduce plasmashell frozen time

jtamate created this revision. jtamate added reviewers: Frameworks, Plasma. Restricted Application added projects: Plasma, Frameworks. Restricted Application added a subscriber: plasma-devel. jtamate requested review of this revision. REVISION SUMMARY Third part of

D10669: Reduce plasmashell frozen time

jtamate created this revision. jtamate added reviewers: Plasma, Frameworks. Restricted Application added projects: Plasma, Frameworks. Restricted Application added a subscriber: plasma-devel. jtamate requested review of this revision. REVISION SUMMARY Second part of

D10629: Avoid a double emitResult()

This revision was automatically updated to reflect the committed changes. Closed by commit R120:4b68374192ac: Avoid a double emitResult() (authored by jtamate). REPOSITORY R120 Plasma Workspace CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10629?vs=27523=27571 REVISION DETAIL

D10629: Avoid a double emitResult()

jtamate added a comment. In D10629#209358 , @broulik wrote: > I did a objdump -d to be sure: > dataengines/places/setupdevicejob.cpp setError is from KJob and it only sets a member value (d->error = errorCode;) >

D10629: Avoid a double emitResult()

jtamate updated this revision to Diff 27523. jtamate added a comment. I haven't found more instances of setResult followed by emitResult. REPOSITORY R120 Plasma Workspace CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10629?vs=27456=27523 REVISION DETAIL

D10627: Reduce plasmashell frozen time to almost nothing

jtamate retitled this revision from "Reduce plasma lock up time to almost nothing" to "Reduce plasmashell frozen time to almost nothing". jtamate edited the summary of this revision. jtamate edited the test plan for this revision. REPOSITORY R120 Plasma Workspace BRANCH polling (branched

D10629: Avoid a double emitResult()

jtamate created this revision. jtamate added reviewers: Frameworks, Plasma. Restricted Application added a project: Plasma. Restricted Application added a subscriber: plasma-devel. jtamate requested review of this revision. REVISION SUMMARY ServiceJob::setResult already does a emitResult.

D10627: Reduce plasma lock up time to almost nothing

jtamate added a comment. May I do the other 2 commits without another revision? REPOSITORY R120 Plasma Workspace BRANCH polling (branched from master) REVISION DETAIL https://phabricator.kde.org/D10627 To: jtamate, #frameworks, #plasma, davidedmundson Cc: davidedmundson,

D10627: Reduce plasma lock up time to almost nothing

jtamate created this revision. jtamate added reviewers: Frameworks, Plasma. Restricted Application added a project: Plasma. Restricted Application added a subscriber: plasma-devel. jtamate requested review of this revision. REVISION SUMMARY CCBUG: 358231 CCBUG: 342056 Even the icon with the

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate added a comment. #10 0x7ff925c93845 in KJob::finishJob(bool) (this=0x221d0d0, emitResult=true) at /g/5kde/frameworks/kcoreaddons/src/lib/jobs/kjob.cpp:101 #11 0x7ff925c93f9b in KJob::emitResult() (this=0x221d0d0) at /g/5kde/frameworks/kcoreaddons/src/lib/jobs/kjob.cpp:300

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate updated this revision to Diff 27444. jtamate added a comment. Now with an Q_ASSERT that really shows the origin of the problem. In the case of pressing the "skip All" button when can't change permissions is: REPOSITORY R244 KCoreAddons CHANGES SINCE LAST UPDATE

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate edited the summary of this revision. REPOSITORY R244 KCoreAddons REVISION DETAIL https://phabricator.kde.org/D9862 To: jtamate, #frameworks, dfaure Cc: michaelh

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate added a reviewer: dfaure. REPOSITORY R244 KCoreAddons REVISION DETAIL https://phabricator.kde.org/D9862 To: jtamate, #frameworks, dfaure Cc: michaelh

D9862: Don't create 2 msgbox (one of them crashes) when can't change rights.

jtamate updated this revision to Diff 27429. jtamate added a comment. I still can trigger the crash without the patch with KF5 5.42 (opensuse) Improved the patch it a little bit, including a QASSERT to know where the double finish originates. REPOSITORY R244 KCoreAddons CHANGES SINCE

D10537: Convert from old connect syntax to new one

jtamate updated this revision to Diff 27238. jtamate edited the test plan for this revision. jtamate added a comment. fixed broulik comment. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10537?vs=27235=27238 REVISION DETAIL

D10537: Convert from old connect syntax to new one

jtamate created this revision. jtamate added reviewers: Frameworks, dfaure. Restricted Application added a project: Frameworks. jtamate requested review of this revision. REVISION SUMMARY Starting plasmashell with callgrind I noticed a line: QObject::connect: No such slot

D10531: [QmlObjectIncubationController] Don't call incubate() while already incubating

jtamate added a comment. Also measured scientifically accurate with a stopwatch 2:34 minutes in callgrind wihtout the patch. I refs: 10,827,698,978 2:20 minutes in callgrind with the patch.I refs: 10,668,473,812 +1 REPOSITORY R242 Plasma Framework (Library)

D9840: Tentative patch to reduce I/O overhead of plasmashell when copying files

jtamate added a comment. In D9840#204587 , @dfaure wrote: > Can you strace vivaldi to see what it's doing? It must be doing something wrong if it's triggering so much I/O. This is related to https://phabricator.kde.org/D10342 I've

D9840: Tentative patch to reduce I/O overhead of plasmashell when copying files

jtamate added a comment. > ! In D9840#203803 , @ngraham wrote: > Do any of the active or recently-landed patches address this? No, but I have one in https://phabricator.kde.org/D10342 REPOSITORY R309 KService REVISION DETAIL

D9840: Tentative patch to reduce I/O overhead of plasmashell when copying files

jtamate added a comment. Just to let you know that I know how to reproduce the high I/O of plasmashell and continuous calls to Kservice. You can reproduce it just having opened the vivaldi web browser. (I never though it could be a problem). If it goes to a webpage that still uses flash,

D10124: Faster simplejob start

This revision was not accepted when it landed; it landed in state "Needs Review". This revision was automatically updated to reflect the committed changes. Closed by commit R241:42fed2e70a21: Faster simplejob start (authored by jtamate). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE

D10124: Faster simplejob start

jtamate updated this revision to Diff 26787. jtamate added a comment. Removed the private slots. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D10124?vs=26239=26787 BRANCH master REVISION DETAIL https://phabricator.kde.org/D10124 AFFECTED FILES

D10358: Repair copying file to VFAT without warnings.

jtamate accepted this revision. jtamate added a comment. This revision is now accepted and ready to land. Even moving to a ntfs filesystem without the rights to change owner doesn't end in an empty file. Very good job. BRANCH silence_vfat_warnings REVISION DETAIL

D10376: kio_file: skip error handling for initial perms during file copy

jtamate accepted this revision. jtamate added a comment. This revision is now accepted and ready to land. Even moving to a ntfs filesystem without the rights to change owner doesn't end in an empty file. Very good job. REPOSITORY R241 KIO BRANCH master REVISION DETAIL

D9983: Don't stat(/etc/localtime) between read() and write() copying files

This revision was automatically updated to reflect the committed changes. Closed by commit R241:8d73867b3d43: Dont stat(/etc/localtime) between read() and write() copying files (authored by jtamate). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE

D10358: Repair copying file to VFAT without warnings.

jtamate requested changes to this revision. jtamate added a comment. This revision now requires changes to proceed. Copying to a VFAT has no problems. Moving to a VFAT has the following problem: An Access Denied Information appears after the file have been created. In both cases, there

D9983: Don't stat(/etc/localtime) between read() and write() copying files

jtamate marked 2 inline comments as done. jtamate added a comment. So, good to go? REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D9983 To: jtamate, #frameworks, dfaure Cc: fvogt, ngraham, michaelh

D10124: Faster simplejob start

jtamate added a reviewer: dfaure. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D10124 To: jtamate, #frameworks, mwolff, dfaure Cc: mwolff, broulik, ngraham, anthonyfieroni, michaelh

D10124: Faster simplejob start

jtamate added a comment. False alarm. :-) Do not know what caused it, but after a complete kdesrc-build, the leak is gone. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D10124 To: jtamate, #frameworks, mwolff Cc: mwolff, broulik, ngraham, anthonyfieroni, michaelh

<    1   2   3   4   5   >