Skip to main content
summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2017-01-27Make Tycho's compiler use configuration from ↵I20170130-2000Mykola Nikishov22-36/+28
.settings/org.eclipse.jdt.core.prefs Change-Id: I60cd95811a1c12a8e961d05a1e4da077fc9a3c30 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2017-01-27Bug 511160 - Run Cleanup action on org.eclipse.equinox.p2.ui to addI20170130-0600I20170129-2000I20170128-2000I20170127-2200I20170127-2000Lars Vogel109-0/+567
missing annotations Change-Id: I53aab78feb00212c8162f992f52b67e4d4dec365 Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2017-01-13Bug 479146 - Adding @Override in InstallWizardY20170126-1700Y20170126-1000Y20170119-1000S4_7_0_M5I20170126-1030I20170126-0505I20170125-2000I20170125-0800I20170124-2000I20170124-0800I20170123-2000I20170123-0830I20170123-0800I20170122-2000I20170121-2000I20170121-0950I20170120-2000I20170119-2000I20170119-1010I20170119-0345I20170118-2000I20170118-1215I20170117-2000I20170116-2000I20170115-2000I20170114-2000I20170113-2000Lars Vogel1-12/+18
Change-Id: I4d5924eec3a175f7826e0027a62e8d2d4aac512e Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2017-01-10Bug 509473: Update version number of simpleconfigurator and p2.engineY20170112-1300Y20170112-1000I20170112-2000I20170112-0245I20170112-0200I20170111-2000I20170111-0200I20170110-2000Markus Keller4-4/+4
2017-01-10Update version number of simpleconfigurator and p2.enginePascal Rapicault4-4/+4
Change-Id: Ide7411f941c6561cba6dd78c6f1c378044056f3d
2017-01-10Bug 508593 - NPE in AutomaticUpdater.sameProfileAndrey Loskutov1-8/+16
Change-Id: Ie87b44f50faeb87518d776de920325451a086a91 Signed-off-by: Andrey Loskutov <loskutov@gmx.de>
2017-01-10Bug 509473: Bump p2.engine and equinox.simpleconfigurator to JavaSE-1.7.Alexander Kurtakov52-261/+502
Enable some java 1.7 and older code style warnings and fix them: * Add missing @Override. * Remove redundant type declarations. * Remove useless non-javadoc comments. Change-Id: If80ea62f65204cad77275e07fee096a276d35c5c Signed-off-by: Alexander Kurtakov <akurtako@redhat.com>
2017-01-10Bug 509473: Bump p2.engine and equinox.simpleconfigurator to JavaSE-1.7.Alexander Kurtakov4-8/+8
Sync project settings with BREE bump in manifest. Change-Id: I93e6d52f91d93c04bf85413a1c348c3c991f8c89 Signed-off-by: Alexander Kurtakov <akurtako@redhat.com>
2017-01-10Bug 495012 - ProvisioningContext#setMetadataRepositories andLars Vogel1-3/+3
setArtifactRepositories should use varargs Change-Id: I917123d4225320c27cde4e7c05473f37d7f7e713 Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2017-01-10Reason for failed assert that the log file doesn't contain messageMykola Nikishov1-2/+2
Provide details on what lines are missing from what log file. Change-Id: I15f0a0ee37feef6470ffbea631e27ee987a9c0e3 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2017-01-10Prefer an enhanced for statement over the basic oneMykola Nikishov1-3/+3
An enhanced for statement makes our intent explicit: when searching for an extension point, we are just iterating over an array of extension points and don't actually need an index variable. Change-Id: Ie328161f639bb1a601e85e3499b3a9eff4643a84 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2017-01-10[jdt settings] Unused local variable is an errorMykola Nikishov1-1/+1
Change-Id: I065dae00209c77fb98fa96e8800849f6b433dc7a Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2017-01-09Bug 508905: Change File.canWrite to Files.isWritable for p2 Droplets.Roland Grunberg8-25/+44
For compatibility with Windows, use java.nio.Files.isWritable(Path) instead of java.io.File.canWrite(). This is due to the fact that canWrite always returns true for directories on Windows. - Enable all tests disabled on Windows by 416b782 - Introduce reallyReadOnly/removeReallyReadonly functions that operate recursively on a directory (like chmod -R on linux) since certain test cases require this Change-Id: I615dfdef60e4abd23c34602bd324c307325e00d4 Signed-off-by: Roland Grunberg <rgrunber@redhat.com>
2017-01-09Bug 509473: Bump p2.engine and equinox.simpleconfigurator to JavaSE-1.7.I20170109-2000Roland Grunberg2-2/+2
org.eclipse.equinox.p2.engine and org.eclipse.equinox.simpleconfigurator should be bumped from a BREE of J2SE-1.5 to JavaSE-1.7 to resolve issues such as Bug 508905. Change-Id: Ib84b397b6c0fb6cbde5a1cf8d3583cb33be5df21 Signed-off-by: Roland Grunberg <rgrunber@redhat.com>
2017-01-09Bug 289518 - Logged backup store error on upgrade/revert of SDKelaskavaia1-8/+12
BackupStore has fullyDelete method. Unfortunately it does not work for links very well If we have lets say 2 files. file1 file2 -> file1 (file2 is link to file1) When this function runs file1 will be deleted, but file2 becomes dead link, then if (!file.exists()) return true; That will return true for file2 even file is not deleted since exists resolve links, so it leaves dead links in directory and whole thingy fails Change-Id: Id78f519907363e1439a35da55cf4174989971f01 Signed-off-by: elaskavaia <elaskavaia.cdt@gmail.com>
2017-01-06Bug 509982 - Version problem withI20170108-2000I20170107-2000Thomas Watson5-4/+15
org.eclipse.equinox.p2.ui.sdk.scheduler Change-Id: I741aa71eff77e83ae94efb3a2405bc7d9a350a12 Signed-off-by: Thomas Watson <tjwatson@us.ibm.com>
2017-01-05Bug 509973 - Comparator errors in I20170105-0320Y20170105-1040Y20170105-1000I20170106-2000I20170105-2000Thomas Watson4-1/+5
Change-Id: I95b8452e1ecb7770f798864b6e7f943e3bce9a12 Signed-off-by: Thomas Watson <tjwatson@us.ibm.com>
2017-01-05Bug 509054 - Consume ECF 3.13.3Alexander Kurtakov2-12/+12
Adjust to ECF source features names change caused by it using autogenerated source feature by tycho. Bump the versions in core feature. Change-Id: I017e831e8bfdab81fb1e382b361645fb9ef32896 Signed-off-by: Alexander Kurtakov <akurtako@redhat.com>
2016-12-16Fix 'always true' assertionY20161229-1000Y20161223-1030Y20161222-1000I20170105-0320I20170105-0230I20170104-2000I20170104-0800I20170103-2000I20170102-2000I20170101-2000I20161231-2000I20161230-2000I20161229-2000I20161228-2000I20161227-2000I20161226-2000I20161225-2000I20161224-2000I20161223-2000I20161223-0615I20161222-2000I20161221-2000I20161221-0400I20161220-2000I20161220-0530Mykola Nikishov4-7/+14
Use org.junit.Assert's assertNotEquals(String, Object, Object) to check if two MD5 are different. junit.framework.TestCase's assertNotSame(String, Object, Object) compares arguments by reference and for String arguments, in most cases, returns true even if both arguments are equal by value. As we need a negative assertion (two MD5 are not equal), junit.framework.TestCase doesn't fit the bill - it does provide assertEquals() but not assertNotEquals(). Change-Id: I1ce6123d9102de963c92da88bea5e713ecdf44e6 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-12-16Ignore Tycho plugins unsupported by M2EI20161219-2000I20161219-0145I20161218-2000I20161217-2000I20161216-2000Mykola Nikishov1-0/+44
Change-Id: I8108e7d9ce3cf596e21223e091b6c24a7364a794 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-12-16Prefer an enhanced for statement over the basic oneMykola Nikishov1-3/+3
An enhanced for statement makes our intent explicit: we are just iterating over an array and don't actually need an index variable. Change-Id: I45705c0fb3d41e2dc3502ea5c285ccbfe8501a66 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-11-23Bug 498116 - Replace fixed check for updates scheduler with a fuzzy oneY20161215-1200Y20161208-1000Y20161201-1000Y20161124-1000S4_7_0_M4I20161215-2000I20161215-0915I20161215-0830I20161215-0730I20161215-0130I20161214-2000I20161214-1005I20161213-2000I20161213-0600I20161213-0300I20161212-2000I20161211-2000I20161210-2000I20161208-0830I20161208-0730I20161207-2000I20161207-1340I20161207-0800I20161207-0515I20161206-2000I20161206-0800I20161205-2000I20161205-0800I20161204-2000I20161203-2000I20161202-2000I20161201-2000I20161201-0800I20161130-2000I20161129-2330I20161129-2000I20161128-2000I20161128-0600I20161128-0330I20161127-2000I20161127-1300I20161126-2000I20161126-0815I20161125-2000I20161124-2000Mikael Barbero12-238/+220
Checking the servers for update at a fixed time on fixed day of the week can kneel the servers. However, activating automatic check for updates on EPP packages is only possible by specifying a fix day and hour (see bug 421779 for the decision). This patch replace the fixed time of the day/week of the day way check for updates are scheduled. Users can choose to check "once a day", "once a week" or "once a month". The initial time used for next checks is the time when the user activates the option. However, it never checks for update exactly after a day, a week or a month. It introduces some randomness: - If the delay for checking for update is passed, it schedules a check sometime in the next 8 hours. - If the delay is well overpassed (see definition of "well overpassed" below), it schedules a check in the next hour. The delay is considered "well overpassed" depending on the recurrence of the check: - If the recurrence is "once a day", the delay is considered overpassed after a day and 6 hours. - If the recurrence is "once a week", the delay is considered overpassed after a week and 2 days. - If the recurrence is "once a month", the delay is considered overpassed after a week and 6 days. It also migrates automatically the fixed weekday/time preferences to the new fuzzy scheduler. Change-Id: I74a8bfba6b711e98eb795c1f924840e68789f102 Signed-off-by: Mikael Barbero <mikael@eclipse.org> Signed-off-by: Pascal Rapicault <pascal@rapicorp.com>
2016-11-13Bug 506432 - [ui] p2 install dialog shouldn't be modalY20161117-1000I20161123-2000I20161122-2000I20161121-2000I20161120-2000I20161119-2000I20161118-2000I20161117-2000I20161116-2000I20161115-2000I20161114-2000I20161114-0355I20161113-2000Lars Vogel1-1/+12
Checks if the dialog is already open and activates the corresponding shell. This way we keep the existing flow of having only one open installation dialog, we we decided is a nicer user flow. Change-Id: I3542e122312ad756011123a4bf1adf3054023fab Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2016-11-08Bug 506432 - [ui] p2 install dialog shouldn't be modal Y20161110-1000I20161112-2000I20161112-0405I20161111-2000I20161111-1220I20161111-0900I20161111-0740I20161111-0510I20161110-2030I20161110-1630I20161110-1225I20161110-1130I20161110-0710I20161110-0510I20161109-2000Lars Vogel1-1/+1
Change-Id: I465a98c8d0c9034c194f7496be3a8efc3e1cc25b Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2016-11-08Bug 493193 - MirrorApplication should explicitly use MD5 artifact comparatorMykola Nikishov2-7/+17
The only place that documents the usage of MD5 comparator in MirrorApplication is Equinox p2 Repository Mirroring wiki page [1]: Adding the argument -comparator <comparator ID> specifies the mirroring application should use an Artifact Comparator with an ID of "comparator ID" to compare artifact descriptors. The mirroring application uses the "MD5 Comparator" to compare the MD5 hash property of the artifact descriptors if no comparator is defined. If no comparator is defined, ArtifactComparatorFactory's getArtifactComparator(String) will be finally called with null as a comparator ID. It's an ArtifactComparatorFactory's implementation detail that it will return the first configured one. MD5 comparator is the only one available now, but this may change and break MirrorApplication's expectation. To avoid such breakage, MirrorApplication declares an explicit dependency on MD5 comparator. So, when '-compare' CLI parameter is true but '-comparator' CLI parameter is missing, it will request MD5 artifact comparator to be used. Also improve testBaselineCompareUsingMD5Comparator() and testCompareUsingMD5Comparator(). These two tests should: - not only set 'compare' to true but also set a desired comparator's id explicitly - run twice, for an empty comparator id and MD5. To not copy-n-paste them, use @Theory [2]. To make theories work, run with Theories test runner. [1] https://wiki.eclipse.org/Equinox_p2_Repository_Mirroring [2] https://github.com/junit-team/junit4/wiki/Theories Change-Id: I5adda382c2824b6488af3716fc645cb6eff14b94 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-11-08Bug 493193 - MirrorApplication should explicitly use MD5 artifact comparatorMykola Nikishov2-5/+48
As test for the fix will use JUnit4 features, don't mix concerns and convert NewMirrorApplicationArtifactTest into a JUnit4 annotation-based test now, in a separate step. Change-Id: Ia0659146bc5c4e951cad9428479c8a593daef51a Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-11-08Revert "Revert "Bug 490028 - Extract MD5 checksum calculation into a ↵I20161108-1015Thomas Watson2-24/+2
separate class"" This reverts commit 6217988b54787e30609f9289ae9ff5b6aa7b33ae.
2016-11-08Revert "Bug 490028 - Extract MD5 checksum calculation into a separate class"Thomas Watson2-2/+24
This reverts commit b85882be2121b5eb9667277da75604f4b2b797d0.
2016-11-07Add tests for IPublisherInfo.A_NO_MD5 contractI20161108-0800Mykola Nikishov1-0/+17
When creating an artifact descriptor with PublisherHelper, createArtifactDescriptor(IPublisherInfo, IArtifactRepository, IArtifactKey, File) will generate MD5 hash if (info == null || (info.getArtifactOptions() & IPublisherInfo.A_NO_MD5) == 0) is true. As current tests cover only 1 out of 4 this condition's branches, add tests for missing branches that assert IPublisherInfo.A_NO_MD5 contract. Change-Id: I9912f01294cee8288a61520086d590f62524fff9 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-11-07Bug 490028 - Extract MD5 checksum calculation into a separate classMykola Nikishov2-24/+2
Since the previous commit [1], RepositoryUtilities' single method just proxies the call down to ChecksumProducer. Get rid of internal RepositoryUtilities completely. [1] 5b2f061cd8ecf96a37783657e7ee7a0110c9d26d Change-Id: Icf962618f1a91c0c5e9f5ec67ff4c587cfaec458 Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-11-04Bug 506432 - [ui] p2 install dialog shouldn't be modal Lars Vogel1-2/+4
Change-Id: Idf9ed2e81c31ce2031c42f8b963119a1df801bfc Signed-off-by: Lars Vogel <Lars.Vogel@vogella.com>
2016-10-31Bug 506597 - Build failure on I20161026-2000Y20161103-1000I20161101-0800Thomas Watson12-12/+12
Change-Id: I424cf3e36674c7529a96f113cf0ef16c8745b3f8 Signed-off-by: Thomas Watson <tjwatson@us.ibm.com>
2016-10-31Bug 483286 - NullPointerException in AbstractRepositoryManager.flushCache()Eike Stepper1-5/+7
Change-Id: I8e010c2a49d8ce75e5d491201800d38c924dfe91 Signed-off-by: Eike Stepper <stepper@esc-net.de>
2016-10-31Rename comparatorPoint constant to follow Java naming conventionsMykola Nikishov1-3/+3
Change-Id: Ic488ddba911af6c076d346fd6f669ec7a04ded6e Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-10-31Bug 506142 - P2 properties dialogue unnecessary escapes & in textelaskavaia1-15/+1
control causing double && Change-Id: I8569368a520de2ce0b5d520ac4a6933d17455bf8 Signed-off-by: elaskavaia <elaskavaia.cdt@gmail.com>
2016-10-27Bug 506008 - query for requirements returns everything if none elaskavaia1-0/+3
- fix - if there is no requirements return null since there is nothing to query Change-Id: I826cac8c60bcce38081c06ad00cd8bee9a60cd62 Signed-off-by: elaskavaia <elaskavaia.cdt@gmail.com>
2016-10-27Bug 506597 - Build failure on I20161026-2000Y20161027-1250Y20161027-1000I20161027-0700Markus Keller6-2/+6
2016-10-17Bug 501505 - Several M2 candidate features decrease versions compared toY20161024-2045Y20161024-1300Y20161024-0700Y20161021-1200Y20161021-0600Y20161020-1800Y20161020-1000Y20161018-1100I20161027-0200I20161026-2000I20161026-1100I20161026-0800I20161025-2000I20161024-2000I20161024-1000I20161024-0410I20161023-2000I20161018-0800Thomas Watson4-4/+4
Neon.1 candidate Change-Id: Ie1c8358aab9d82acd4cbad6260132ba5727e716e Signed-off-by: Thomas Watson <tjwatson@us.ibm.com>
2016-10-04Bug 502515 - Neon still reports a p2 header of "p2 mars-sr0"Y20161013-1000Y20161006-1000I20161011-0800I20161005-1430I20161005-1045I20161005-0815I20161004-1000Pascal Rapicault3-3/+5
Change-Id: I28549f765f70cd3a0d78a21fbc0469f6c6d9ab8e Signed-off-by: Pascal Rapicault <pascal@rapicorp.com>
2016-10-04Bug 495726: MetadataRepositoryElementWrapper ignores parent contextI20161004-0800elaskavaia1-5/+9
If have content set for MetadataRepositories and children of it is MetedataRepository, because of this wrapper context is not passed around, for example if I say I don't want filters, it will only apply to root and not individual repos which does not make any sense Change-Id: I8a89d9366f9cd7575f36c85c66e025b782d6019a Signed-off-by: elaskavaia <elaskavaia.cdt@gmail.com>
2016-09-27Bug 495726 - API: InstallWizardPage ui elements do not respect policy Alena Laskavaia9-17/+37
P2 UI Wizard pages do not respect provisioning ui which is passed to them when it comes to querying elements, because IUElementListRoot which is the root of all elements does not extend QueriedElement so getProvisioningUI() on the elments defaults to ProvisioningUI.getDefaultUI() Fix is to add an API to accept ProvisioningUI as an argument. All places that were creating IUElementListRoot now pass appropriate ui to the constructor. For backward compatibility if ui is not set or set to null it will return default UI as before. Change-Id: I1d8d87339812c874eeee7127e762731b600cb4bf Signed-off-by: Alena Laskavaia <elaskavaia.cdt@gmail.com>
2016-09-19Bug 499623: Do not run testLoadConfigurationExtended on Windows.Y20160929-1000Y20160922-1000I20160927-0800I20160920-0800Roland Grunberg1-0/+6
Currently, test utility methods do not guarantee a folder can be made read-only and asserted using java.io.File.canWrite(). As a result extended configuration tests will not work correctly on Windows. Change-Id: I2e4f320587aea45c22a4b95b57f6817c3e8a09ff Signed-off-by: Roland Grunberg <rgrunber@redhat.com>
2016-08-30Bug 461102 - NPE in OperationFactory.listInstalledElementsY20160915-1000Y20160908-1000Y20160901-1000I20160915-0230I20160915-0200I20160914-2300I20160914-2000I20160914-0800I20160914-0430I20160914-0030I20160913-2000I20160913-0800I20160912-2000I20160912-1005I20160912-0800I20160912-0430I20160912-0225I20160912-0140I20160911-2015I20160906-0800Sergey Prigogin1-2/+6
Change-Id: I3bfff65bdcd579398434c1a19c1642be5d2f34c0 Signed-off-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
2016-08-30Bug 371970 - NPE atSergey Prigogin1-4/+20
org.eclipse.equinox.internal.p2.engine.SimpleProfileRegistry.lockProfile Change-Id: I789887493f8b5e3b601a6b00c8ec7928c77e85e3 Signed-off-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
2016-08-24Bug 500225: Endless loop in Install/Update dialogY20160825-1000I20160830-0800I20160824-1429Markus Keller1-1/+1
2016-08-24fixed Javadoc errorsMarkus Keller3-5/+5
2016-08-17Bug 488504 - [ui] Allow the filter for available IUs to also match on their ↵Y20160818-1000I20160823-1359I20160823-0759Nitin Dahyabhai8-15/+87
description Signed-off-by: Nitin Dahyabhai <nitind@us.ibm.com> Change-Id: I19d173c5fdad89705d966c954343eda607409f46
2016-08-17Bug 494822 - Pass a range specification unconditionallyMykola Nikishov1-6/+2
When sending request for transfer of a remote file to local file storage from FileReader.sendRetrieveRequest(URI, OutputStream, DownloadRange, boolean, IProgressMonitor), drop redundant null check for DownloadRange parameter and pass it to IRetrieveFileTransferContainerAdapter's sendRetrieveRequest(IFileID, IFileRangeSpecification, IFileTransferListener, Map) as is because a range specification for retrieving a portion of the given remote file is allowed to be null: If null the entire file will be retrieved (as per sendRetrieveRequest(IFileID, IFileTransferListener, Map). Bug: 266243 Change-Id: If957ebd23e4a229db873cb6d4df3e10d5b93c75b Signed-off-by: Mykola Nikishov <mn@mn.com.ua>
2016-08-17Bug 482765 - NullPointerException in ↵Carsten Reckord1-2/+2
TextSearchControl.createLabelButtonControl(198) Changed the broken icon path to point to the png version. Fall back to "missing image" descriptor when the provided descriptor is null. Change-Id: I3fb3cdd08ea7413f388661f2657070710463620e Signed-off-by: Carsten Reckord <reckord@yatta.de>
2016-08-12Bug 499623 - Support extended configurations in SimpleConfigManipulator.Roland Grunberg2-3/+64
SimpleConfiguratorManipulatorImpl#loadConfiguration(BundleContext, String) only returns the list of BundleInfo elements from the platform itself. Extend this mechanism to also return extended BundleInfo elements (when p2.fragments property is set). Change-Id: I2230aff75bb20f8e38dd8a7a22c4cba6aefd6df8 Signed-off-by: Roland Grunberg <rgrunber@redhat.com>

Back to the top