T10262: Integrate KIO Slaves into file system using FUSE gateway

2019-03-13 Thread Alexander Saoutkin
feverfew added a comment. @ngraham I'll be having some free time this summer. I was thinking if we could solidify this task into exactly what we need to do so we can get a proposal in for GSoC? TASK DETAIL https://phabricator.kde.org/T10262 To: cfeck, feverfew Cc: feverfew, #frameworks,

D23194: Making FileJob behave consistently.

2019-08-16 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > chinmoyr wrote in file.cpp:520 > A loop is required here. The docs don't really specify anything about reading > data in one go. Exactly. The docs weren't specific at all. smb/sftp would read once, and return the result. the file slave would

D23194: Making FileJob behave consistently.

2019-08-16 Thread Alexander Saoutkin
feverfew updated this revision to Diff 63861. feverfew added a comment. Addressing comments REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23194?vs=63841=63861 BRANCH ConsistentRead (branched from master) REVISION DETAIL

D23159: Prevent error() being emitted when purposefully reading 0 bytes

2019-08-14 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > apol wrote in file.cpp:520 > this should be > > { > Q_EMIT data({});` > return; > } > > I'm not sure that we'd need an empty data emitted then. Are you trying to fix > a specific bug that triggers this? I('ve not seen any

D23207: Fixing implementation of FileJob interface in smb/sftp slaves

2019-08-16 Thread Alexander Saoutkin
feverfew updated this revision to Diff 63876. feverfew added a comment. Unnecessary min version upgraded REPOSITORY R320 KIO Extras CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23207?vs=63875=63876 BRANCH fixSFTP (branched from master) REVISION DETAIL

D23194: Making FileJob behave consistently.

2019-08-16 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > chinmoyr wrote in file.h:117 > Nitpick; why not just closeFile() ? That's how sftp does it, so for consistency. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D23194 To: feverfew, dfaure, fvogt, chinmoyr, apol Cc:

D23207: Fixing finished() called after error() in smb slave. Do not call finished() in open/read/write/seek operations

2019-08-16 Thread Alexander Saoutkin
feverfew created this revision. Herald added projects: Dolphin, Frameworks. Herald added subscribers: kfm-devel, kde-frameworks-devel. feverfew requested review of this revision. REPOSITORY R320 KIO Extras BRANCH fixSFTP (branched from master) REVISION DETAIL

D23194: Making FileJob behave consistently.

2019-08-16 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > file.cpp:550 > +// Make sure data gets to disk. > +if (mFile->flush()) > +written(bytesWritten); Flushing on every write IMO is a bit expensive, but I don't like the fact that close() can't report errors and so we

D23194: Making FileJob behave consistently.

2019-08-16 Thread Alexander Saoutkin
feverfew marked 4 inline comments as done. feverfew added inline comments. INLINE COMMENTS > chinmoyr wrote in filejob.h:48 > Document the methods in SlaveBase as well. I've used @see to help me out. Is that enough? I don't want to duplicate the docs, especially if someone changes it and

D23207: Fixing implementation of FileJob interface in smb/sftp slaves

2019-08-16 Thread Alexander Saoutkin
feverfew retitled this revision from "Fixing finished() called after error() in smb slave. Do not call finished() in open/read/write/seek operations" to "Fixing implementation of FileJob interface in smb/sftp slaves". feverfew edited the summary of this revision. feverfew edited the test plan

D23159: Prevent error() being emitted when purposefully reading 0 bytes

2019-08-15 Thread Alexander Saoutkin
feverfew abandoned this revision. feverfew added inline comments. INLINE COMMENTS > feverfew wrote in file.cpp:524 > I think that would be appropriate for a separate revision which I'm doing. > This revision is only to handle the case outlined in the summary. Actually I've realised using the

D23194: Making FileJob behave consistently.

2019-08-15 Thread Alexander Saoutkin
feverfew edited the summary of this revision. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D23194 To: feverfew, dfaure, fvogt, chinmoyr, apol Cc: kde-frameworks-devel, LeGast00n, michaelh, ngraham, bruns

D23194: Making FileJob behave consistently.

2019-08-15 Thread Alexander Saoutkin
feverfew edited the summary of this revision. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D23194 To: feverfew, dfaure, fvogt, chinmoyr, apol Cc: kde-frameworks-devel, LeGast00n, michaelh, ngraham, bruns

D23194: Making FileJob behave consistently.

2019-08-15 Thread Alexander Saoutkin
feverfew created this revision. feverfew added reviewers: dfaure, fvogt, chinmoyr, apol. Herald added a project: Frameworks. Herald added a subscriber: kde-frameworks-devel. feverfew requested review of this revision. REVISION SUMMARY This patch does the following: 1. Makes sure the

D23159: Prevent error() being emitted when purposefully reading 0 bytes

2019-08-15 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > chinmoyr wrote in file.cpp:524 > This doesn't support error reporting. Why not refactor this loop using > read(char *, qint64)? I believe you can accommodate your changes inside the > loop cleanly then. I think that would be appropriate for a

D23194: Making FileJob behave consistently.

2019-08-19 Thread Alexander Saoutkin
feverfew updated this revision to Diff 64051. feverfew added a comment. Making comments clearer REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23194?vs=63861=64051 BRANCH ConsistentRead (branched from master) REVISION DETAIL

D23159: Prevent error() being emitted when purposefully reading 0 bytes

2019-08-14 Thread Alexander Saoutkin
feverfew created this revision. feverfew added reviewers: dfaure, fvogt, chinmoyr. Herald added a project: Frameworks. Herald added a subscriber: kde-frameworks-devel. feverfew requested review of this revision. REVISION SUMMARY Currently when 0 is passed as a parameter, if the file to be read

D22708: [WIP] Add a kded module infra for fuse mount services

2019-08-21 Thread Alexander Saoutkin
feverfew added a comment. In D22708#516128 , @ngraham wrote: > Oh good. :) Is there anywhere I can follow the progress? All my work goes on KIOFuse goes into review here: https://gitlab.com/Vogtinator/kio-fuse/merge_requests

D22708: [WIP] Add a kded module infra for fuse mount services

2019-08-21 Thread Alexander Saoutkin
feverfew added a comment. In D22708#516132 , @ngraham wrote: > In D22708#516131 , @feverfew wrote: > > > In D22708#516128 , @ngraham wrote: > > > > >

D22708: [WIP] Add a kded module infra for fuse mount services

2019-08-21 Thread Alexander Saoutkin
feverfew added a comment. In D22708#507449 , @ngraham wrote: > What? Don't worry, all is fine. I've taken over Chinmoy's work on this for now. We agreed on a better way to go about integrating KIOFuse into KIO. REPOSITORY R241 KIO

D23207: Fixing implementation of FileJob interface in smb/sftp slaves

2019-08-26 Thread Alexander Saoutkin
feverfew added reviewers: sitter, dfaure. REPOSITORY R320 KIO Extras REVISION DETAIL https://phabricator.kde.org/D23207 To: feverfew, chinmoyr, fvogt, sitter, dfaure Cc: kde-frameworks-devel, kfm-devel, aprcela, vmarinescu, fprice, LeGast00n, MrPepe, fbampaloukas, alexde, GB_2, Codezela,

D23207: Fixing implementation of FileJob interface in smb/sftp slaves

2019-08-30 Thread Alexander Saoutkin
This revision was automatically updated to reflect the committed changes. Closed by commit R320:28fdc41f1d45: Fixing implementation of FileJob interface in smb/sftp slaves (authored by feverfew). REPOSITORY R320 KIO Extras CHANGES SINCE LAST UPDATE

D23194: Making FileJob behave consistently.

2019-08-30 Thread Alexander Saoutkin
This revision was automatically updated to reflect the committed changes. Closed by commit R241:3154fc3e096e: Making FileJob behave consistently. (authored by feverfew). REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23194?vs=64673=64976 REVISION DETAIL

D23384: [WIP] Adding support for mounting KIOFuse URLs for applications that don't use KIO

2019-08-23 Thread Alexander Saoutkin
feverfew created this revision. feverfew added reviewers: fvogt, chinmoyr. Herald added a project: Frameworks. Herald added a subscriber: kde-frameworks-devel. feverfew requested review of this revision. TEST PLAN Make sure kded module is installed and loaded. Use dolphin and observe that

D23454: Fixing bug where MTP slave does not return error in stat()/mimetype()

2019-08-25 Thread Alexander Saoutkin
feverfew created this revision. feverfew added reviewers: chinmoyr, akrutzler, elvisangelaccio, dfaure. Herald added projects: Dolphin, Frameworks. Herald added subscribers: kfm-devel, kde-frameworks-devel. feverfew requested review of this revision. REVISION SUMMARY This patch fixes two cases

D23194: Making FileJob behave consistently.

2019-08-25 Thread Alexander Saoutkin
feverfew added inline comments. INLINE COMMENTS > dfaure wrote in filejob.h:112 > Why would someone request reading 0 bytes? That doesn't seem sensible to me. Well it can happen and this is technically more correct than saying that an empty `QByteArray()` is always EOD. In fact, the one of the

D23194: Making FileJob behave consistently.

2019-08-26 Thread Alexander Saoutkin
feverfew marked an inline comment as done. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D23194 To: feverfew, dfaure, fvogt, chinmoyr, apol Cc: kde-frameworks-devel, LeGast00n, GB_2, michaelh, ngraham, bruns

D23194: Making FileJob behave consistently.

2019-08-26 Thread Alexander Saoutkin
feverfew marked 3 inline comments as done. REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D23194 To: feverfew, dfaure, fvogt, chinmoyr, apol Cc: kde-frameworks-devel, LeGast00n, GB_2, michaelh, ngraham, bruns

D23194: Making FileJob behave consistently.

2019-08-26 Thread Alexander Saoutkin
feverfew updated this revision to Diff 64672. feverfew added a comment. - Fixing typos - Adding test for reading of 0 bytes and asserting that the close signal is actually emitted. REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23194?vs=64051=64672

D23194: Making FileJob behave consistently.

2019-08-26 Thread Alexander Saoutkin
feverfew updated this revision to Diff 64673. feverfew added a comment. - Removing flush on write REPOSITORY R241 KIO CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D23194?vs=64672=64673 BRANCH ConsistentRead (branched from master) REVISION DETAIL

D23207: Fixing implementation of FileJob interface in smb/sftp slaves

2019-08-27 Thread Alexander Saoutkin
feverfew added a comment. In D23207#520233 , @sitter wrote: > Not knowing the background here at a glance I would argue that SlaveBase in KIO should be getting state verification on all of this,. Sorry, I'm not sure what you mean by

D24076: add a simple smoke test for slaves by using kioclient5

2019-09-19 Thread Alexander Saoutkin
feverfew added a comment. I won't review, but would like to note I think that with KIOFuse, in conjuction with fio we can do some really complex testing on KIO, correct me if I'm wrong. If I'm right, then we can test KIO nicely, although if we do notice a