topic: don't abort if there is no current topic in stack() revset Named branches support has been added and we should not abort if there is no active topic.

topics: return a parse error if stack() revset is passed with argument

tests: add test to show topics working with shelve

topics: update current topic to the topic of newly rebased commit (issue5551) The rebase code passes branchmerge equals to True while updating to the rebased commit. We need to make sure topic is preserved even after rebase and hence we need to update the topic even when branchmerge argument is set to True. But there is a twist in the tale, merge also uses this part of code and we allow to update topic when brancmerge is True, in merge cases the topic after merge will the topic of the destination commit, not the topic of working directory parent. So we need the function to have information about whether a rebase is going on, and we do it by wrapping the rebase command and storing some value in the config. This is a bit hacky but works for now. This patch fixes issue related to loosing of topic while rebase. Thanks to Boris Feld for the rigourous tests.

docs: some fixes to the help text

topics: pass 'topics' to ui.formatter instead of 'bookmarks'

topic: also insert the extra head check with using the new head checking This will get the protection in place in all cases.

topic: some document for an obscure function

topic: adjust head checking wrapping to not interfere with concurrent push In the previous form, the new topic head was detected as affected and the push was rejected. We introduce a of 'test-topic-push.t' using the new race checker as there have been many compatibility issue with this new checkers.

topic: also detect head checking using the concurrent part

test: avoid 'test-check-commit.t' complaining copied code We switch to git diff.

topic: small test cleanuo The section was repeated.

topic: move the heads data to the branch line when appropriates The topic line is dropped as it is quite confusing.

topic: add some initial support for using stack on named branch Stack is a useful command that can make sense in other context. The current support is hacky and awful, but we have something!

topic: explicitly pass topic as a keyword argument This clarify all callers before adding more logic related to bare branch in stack.

topic: makes split keep the topic Refactor split to extract the preparation of the working directory in a separate function. This helps topic wrap this function to save topic before update and restore it after preparing the working directory just before the split.

topic: add tests for fold behavior with topics Clarify the fold behavior for keeping topics.

topics: pass quietempty=True to suppress useless output while changing topics This will help in suppressing the output "0 files updated, 0 file removed, ..."

topics: return early so that other if conditions don't get execute (issue5600) We were not returning after the _changetopics() call and hence the if's below it were also executed which resulted in changing of current topic. This prevent current topic to be updated even when we only affect changeset unrelated to the working copy. Not that the rebase output change as it expose a new bug (rebased changeset gain a topic)

test: extra output to understand the next changeset better

topics: move the logic to change or clear current topic into a new function It adds a new _changecurrenttopic which changes or clears the current topic.

tests: expand output to check topic activation This will help understanding later changesets.

branching: merge stable into default

split: preserve author of the original changeset in the successors Previously, the new changesets were authored by the default user.

grab: properly quote hg executable in the alias Breakage has been reported on Windows.

topic: improve the revset used to return name->nodes mapping The 'topic' version already filters public changeset out, and skipping 'changectx' creation will helps performance.

topic: update the changectx method to respect phases The topic is no longer returned once the changeset is no longer mutable (unless explicitly requested).

topic: avoid crash when topic is loaded but not enabled for a repository

compat: update test to match core default state This output changed in 560ceb654180.

topicmap: massive rework Massively rework the way we build and use topicmap. This bring massive performance benefit. Topic map use to be a fully independant thing that we would switch on and off globaly. The caching on disk was broken so the performance were atrocious. Intead, now the topic are inherited from the 'immutable' map. We gave up on storing them on disk for now since the mutable set is usually small enough. The activation is done by hacking new "filter" on the repository and detection when they are one. This is hacky but core is hard to wrap here. Overall this whole wrapping is really scary and we should massage core API to help it.

topic: use the 'topic' revset in namespace This revset performance just got improved.

topic-revset: update the revset to no longer build changectx Atribute access is expensive, and changectx are very expensive to build, so we skip these for better performance. Before: ! wall 0.012195 comb 0.020000 user 0.020000 sys 0.000000 (best of 217) After: ! wall 0.008816 comb 0.010000 user 0.010000 sys 0.000000 (best of 303) (Before this changeset parent) ! wall 0.213261 comb 0.210000 user 0.210000 sys 0.000000 (best of 45)

topic-revset: changectx creation in the revset The current mutability filter is very expensive, we use a more direct and efficent way to do so: Before: ! wall 0.213261 comb 0.210000 user 0.210000 sys 0.000000 (best of 45) After: ! wall 0.012195 comb 0.020000 user 0.020000 sys 0.000000 (best of 217)

topics: add some noise to rewrittent changeset to prevent hash cycle If we have a changeset with topic `x`, we change it's topic to `y`, fine. When we change it's topic back again to `x`, we get the hash of the the obsoleted changeset which had the topic `x` initially. The same happens for few more cases like clearing the topic of a changeset which initially had no topic. This approach is influenced from cmdutil.amend and other commands (rebase, histedit, etc…)

topic-change: update the working copy along when changing topic of '.' This avoids leaving the working copy behind, removing another large parts of issue5441.

topic-change: cleanup the locking and transaction mechanism The previous code did not garanted we would release all lock (and was a bit more complicated than needed.

topics: use a dict for successors of changesets produced by topic change This patch adds a dictionary to store the successors of changesets which got obsoleted when we change it's topic. This is an improvement in the fix for issue5441.

topic: further simplify the clear logic We just set 'topic' to None and the rest should follow.

topics: drop the clean argument from _changetopics() After this commit, None will represent that we want to clean the topic.

topics: rename '--change' flag to '--rev' flag The --change flag was used to read a revset of which topic should be changed. In mercurial we use --rev for this type of flag. Now things will work as follows: `hg topics topicname`: It will set the current topic `hg topics topicname --rev revset`: Change topic of all the revs in the revset to topicname Further patches will try to achieve the following: `hg topics --clear --rev revset`: clear topic from all the revisions in the revset `hg topics --clear`: clear the current topic (Current behavior) `hg topics --rev revset`: show topics on the revisions in the revset

topics: factor out the logic to change topic in a new function It will help to refactor and fix bugs. Moreover we can re-use the logic.

obslog: update the readme This is release not worthy

obslog: add a comment about the final new line of descriptions

obslog: also display description patch with --patch When activating the --patch option, also display the description patch.

obslog: extra patch indentation code The code is simple and we need to reuse it.

obslog: add a patch option Add support for '--patch' option in obslog for a similar effect than "hg log --patch". The patch support is only limited to a few basic cases that will be extended in the future.

obslog: clarify some sorting code I'm not sure why Boris did it, but this belong to the next patch.

obslog: pass directly unfiltered_repo _debugobshistorydisplaynode used to create two times an unfiltered repo, makes _debugobshistoryrevs create only one unfiltered repo and pass it directly.

obslog: remove useless repo arg _debugobshistorydisplaymarker used to takes repo as an argument but it isn't needed anymore.

obslog: small renaming of _debugobshistorysingle Rename _debugobshistorysingle to _debugobshistoryrevs in order to better reflect its purpose.

refactoring: directly use 'ctx.unstable()' in showstack function This is simpler and more efficient.

packaging: flag default as version 6.5.0.dev This helps to reduce confusion when installing intermediate version.

merge: get stable change back into default

obsfate: improve obsfate output Improve obsfate output: * Remove newline and replace it by '; ' * Improve the hybrid object for making join(obsfate, '|') working.

packagin: mark stable branch as 6.4.1.dev This help prevent confusion with installing intermediate version

topics: abort if user wants to show the stack of a non-existent topic

stack: show unstable state for the current revision if it is one (issue5553) Before this patch if the current revision is unstable, hg stack does not show the unstable state. This patch fixes that by showing both current and unstable state for the current revision if that's unstable.

topics: some minute fixes to the documentation which shows up in help I am not a native but these changes are simple and obvious.

topics: make sure we commit on new parents while changing topics (issue5441) While changing topics of a set of linear commits, we used to commit our new changesets with new topic on parents of its predecessor i.e. changeset before the topic change. If the topic of parent was also changed, that parent will become obsolete and hence resulting the cnew commit in unstable state. For a set of linear commits this repeats and we end up in a tree state as mentioned in the bug. This patch fixes the bug by checking whether the parent was obsoleted and if yes, commit on the new parent.

tests: add a test to show issue5441 in test-topic.t The issue will be fixed in the next patch and the test will be helpful to show that.

merge back with stable (test change from stable has been reverted)

Added tag 6.4.0 for changeset e60248f26f92

packaging: prepare version 6.4.0

readme: fix readme for future 6.4.0

test-compat: merge with mercurial-3.9

test-compat: merge with mercurial-4.0

test-compat: merge with mercurial-4.1

compat-test: merge with stable

test-compat: fix previous merge The merge was a bit too heavy handed in dropping some part of the test.

merge with default Integrate new development around obsfate.

obsfate: by default only display the username if differ from the current one This will will keep the output simple in the simple case.

obsfate: display more data about direct prune markers This does not handle all cases, but this should handle the most common ones.

obsfate: split markers fetch from successor set annotation This will help to improve support for pruned changesets.

obsfate: mark successorsetallmarkers public We are about to use it in another module.

obsfate: mark 'preparesuccessorset' as public It is used in another module so it is not really private.

obsfate: use 'split' as verb when appropriate

obsfate: improve prune support in _successorsetverb In the prune case, still have markers (at least the prune markers). If we want to be able to return these markers we need the underlying function to support it.

obsfate: rename 'successorsset' to 'sset' in the loop The old name make it too close to 'successorssets', and the shorter name is not ambiguous in the context anyway.

template: use hex node in the obsmarkers used in the obsfate template This was pointed by Yuya and yuya is right.

template: use hex successors in obsfate This was spotted by Yuya.

template: use hex-node in successors AS pointed by yuya on the list, the raw data for the template should be hex and not binary.

template: update successors doc I'm not very happy about it, but at lease there is a show description now

template: fix precursors documentation

template: used hex-node for the precursors underlying data As pointed by yuya on mercurial-dev we should not use binary node here.

compat-test: merge with mercurial-3.9

test-compat: merge with mercurial-4.0 branch Minor merge conflict.

compat-test: merge with mercurial-4.1

compat-test: merge with stable * Blackbox format seems to have change when displaying config CLI options. * As obsfate template is now only compatible with hg 4.2+, remove all references to it in test files.

merge with stable

changelog: announce that obsfate is hg > 4.2 now Update changelog to announce the fact that obsfate has been improved and is now only compatible with mercurial version greater than 4.2.

compat: make obshistory._getdifflines compatible with mercurial 4.1 On Mercurial version 4.1 and lower, patch.diff seems to returns an empty list when there is no change. Mercurial 4.2 returns None so we add a if for this case.

obsfate: add a new obsfate template The obsfate template display for each obsolete changeset a line summarizing what changed between the changeset and its successors. This dict is computed in obshistory._preparesuccessorset. It uses obshistory.FORMATSSETSFUNCTIONS which is a list of function that individually compute a part of each dict. You can override fields or add new ones by adding your own function in this list. The format of obsfate is computed in templatekw.obsfatedefaulttempl and can be wrapped if necessary, the code is not quite extendable for the moment but can be refactored later.

obsfate: prepare fatelog tests Prepare obsfate venue by adding new blank output, it will helps clarify which are the new output due to obsfate or not.

obsfate: build more advanced markers for tests Prepare obsfate venue by building more advanced markers by varying obs markers users and dates.

readme: fix entry for 6.4.0 some data were missing and some were in the wrong spot

test: drop now useless config about effect flag The experiment is on by default now.

effectflag: activate effect flag by default Remove forcing effect-flag in test files

effectflag: activate effect flag on test-evolve-obshistory test file Prepare activation of effect flag by default by activating it on test-evolve- obshistory.t test file first.

effectflag: activate effect flag on test-wireproto.t test file Prepare activation of effect flag by default by activating it on test- wireproto.t test file first.

effectflag: activate effect flag on test-uncommit.t test file Prepare activation of effect flag by default by activating it on test- uncommit.t test file first.

effectflag: activate effect flag on test-stabilize-order.t test file Prepare activation of effect flag by default by activating it on test- stabilize-order.t test file first.

effectflag: activate effect flag on test-obsconvert.t test file Prepare activation of effect flag by default by activating it on test- obsconvert.t test file first.

effectflag: activate effect flag for obsmarkers-case-*.t test files Prepare activation of effect flag by default by activating it on test- exchange-obsmarkers-case-*.t test file first.

effectflag: activate effect flag on test-drop.t test file Prepare activation of effect flag by default by activating it on test-drop.t test file first.

effectflag: activate effect flag on test-amend.t test file Prepare activation of effect flag by default by activating it on test- amend.t test file first.

effectflag: activate effect flag on test-stabilize-result.t test file Prepare activation of effect flag by default by activating it on test- stabilize-result.t test file first.

effectflag: activate effect flag on test-split.t test file Prepare activation of effect flag by default by activating it on test-split.t test file first.

effectflag: activate effect flag on test-evolve-cycles.t test file Prepare activation of effect flag by default by activating it on test-evolve- cycles.t test file first.

effectflag: activate effect flag on test-evolve-obshistory-complex.t test file Prepare activation of effect flag by default by activating it on test-evolve- obshistory-complex.t test file first.

effectflag: activate effect flag on test-prune.t test file Prepare activation of effect flag by default by activating it on test-prune.t test file first.

effectflag: activate effect flag on test-evolve.t test file Prepare activation of effect flag by default by activating it on test-evolve.t test file first.

effectflag: activate effect flag on test-obsolete test file Prepare activation of effect flag by default by activating it on test- obsolete.t test file first.

obscache: directly allocate zeroed bytearray This is much faster that incrementally appending to it. On mozilla central without obsolescence markers this move full cache building from 0.3s to nothing visible.

obscache: plug on core 'update cache' method if available I was planning to have this implemented in core but since things are not getting in I'm updating evolve.

obscache: pre-indent code This will make the next changeset clearer.

fix: fix _headssummary call to orig 65cf338258d2 one of two orig call wasn't compatible with mercurial version < 4.3, fix it.

test-update: update hg update output in test files Following hg revision 1b5c61d38a52, hg update now show the commit to which we updated in case of multiple heads. test-obsolete.t and test-sharing.t falls in these situations, acknowledge the new outputted line.

test-drop: all obsmarkers are now dropped Previously, hg drop may leave some obsmarkers behind, like with temporary amend commit. Following mercurial revision 71eb6a098315, strip now correctly remove the associated obsmarkers and as drop is also calling strip, the bug is now resolved.

test-evolve: garbage collection changeset only, not obsmarkers

test-inhibit: adapt to 4.3 version 4.3 introduce markers stripping with strip. We adapt the test output to the (harmless) consequence.

test-exchange-case-D3: avoid stripping obsolescence marker The test needs to have changeset unknown locally with obsolescence markers still applying to them. So we make sure the markers are not stripped (future 4.3 strip markers by default).

test-exchange-case-D2: avoid stripping obsolescence marker The test needs to have changeset unknown locally with obsolescence markers still applying to them. So we make sure the markers are not stripped (future 4.3 strip markers by default).

test-exchange-case-D1: avoid stripping obsolescence marker The test needs to have changeset unknown locally with obsolescence markers still applying to them. So we make sure the markers are not stripped (future 4.3 strip markers by default).

test: fix test-discovery-obshashrange.t new output Update test-discovery-obshashrange.t test file with new mercurial dev output.

fix: fix _headssummary api _headssummary now takes directly the pushop, update the topic extension wrapping to fix the API.

url: update url to point to https://www.mercurial-scm.org/repo/evolve

doc: document the evolve templates

test: fix template condition for obsfate

obscache: add a missing "new line" at the end of a debug message

default: merge back with stable

obscache: shortcut the attribute access for testing The attribute access adds a significant overhead. we can gain -20% performance by doing so.

obshashrange: add an option to protect from accidental activation The new 'obshashrange.max-revs' option is meant to help large installation like bitbucket to avoid impact on repositories that cannot handle it.

obshashrange: basic proof reading of the documentation Nothing major but this should be a bit better.

proofread

effectflag: fix content change detection for filtered revs In some cases (like amended an obsolete changeset), computing the diff and comparing them need to work with filtered revisions. We need to use unfiltered change contexts to safely compute diffs.

debugobsstorestat: adds static about actually obsolete changeses

evolve: fix typo: -list -> --list

Added tag 6.3.1 for changeset e358c0263e46

prepare release 6.3.1

merge with stable

packaging: bump to 6.4.0.dev

serveronly: also backport the fixes from 5adb8bdb935e The fixes to the computation of "relevant" markers. Was missing for server only.

merge stable back into default

readme: merge 6.2.2 and 6.3.0 changelog 6.2.2 never happened.

packaging: mark as version 6.3.1.dev This reduce confusion when installing version from source.

readme: clarify the branch policy

Added tag 6.3.0 for changeset 6da4ca7b3e4f

packaging: prepare release 6.3.0

test-compat: merge with mercurial-3.9

test-compat: merge with mercurial-4.0 branch The new 'olog' command is not working with 3.9 so we skill all mentions of it in the tests. This also apply to the topic extensions.

test-compat: merge with mercurial-4.1 hg log has no trouble field in mercurial 4.0, output fixed.

test-compat: merge with stable Nothing special to report, minor test-metaedit.t output change.

merge with default

compat: fix stablerange for mercurial 3.9 Lrudict.get change of api between mercurial 3.9 and mercurial 4.2. Bypass this limitation by using lrudict.__getitem__ and catch the KeyError directly.

template: fix successors and precursors templates In mercurial 4.1-, the gen argument is expected to be an iterator and not an iterable, fix successors and precursors templates.

lint: remove double newline in test file Test file test-stablesort-criss-cross.t contained two double newline, I remove them.

template: fix precursors template for older mercurial version In mercurial 4.1 and below templatekw._hybrid require the gen argument to be different than None. Manually compute the same default output for these older mercurial versions.

compatibility: backport mercurial 176d1a0ce385 Backport obsstore.relevantmarkers change for including previously forgotten pruned obs markers. Target mercurial version lesser than 4.3. This impact multiple tests. All test changes has been manually checked. Since the computation of relevant obsmarkers have changed, we are bumping the cache schema version. So that caches get recomputed.

evolve: fix import order

compat: adapt to buildobsmarkerspart move The function has been moved in e70d6dbde713 so we must we search in both location.

effetflag: detect other meta (extra) changes Check other changeset meta other than branch and *_source and if there is a change, set the METACHANGED effect flag. Add a test for this scenario with a topic change.

obsdiscovery: add an option to disable all obsdiscovery On large repository, computing the obshashtree can we as slow as resending everything. We give the option for people to opt out.

effectflag: better diff detection Improve the diff detection algorithm to have less false positive like in the case of a rebase that move the hunk in the file but the hunk stays the same.

effectflag: add a test for content detection Add a test for diff detection when the diff is different but only because of line numbers. The test is a append at the end of the file and we try to rebase it. We shouldn't detect a content change, it will be done in the next patch.

test: fix test-evolve-effectflags dates Mercurial < 4.3 don't have default date so add globs in the test-metaedit.t file.

test: fix metaedit tests by forcing dates Mercurial < 4.3 don't have default date so restore aliases forcing date in the test-metaedit.t file.

test: split test-stablesort.t in two This is the longest running test so we split it in two to help with parallel execution.

obsdiscovery: automatically pick protocol in a list This makes the code cleaner, clearer and more flexible. This also prepare the ability to disable obsmarkers discover entirely.

obsdiscovery: extract push-obshashtree discovery in a function This will help use to simplify the code.

obsdiscovery: directly use the 'only' revset for obshashtree The revset now exists and is fast. This remove the needs to hold a reference to "revs" in that code.

obsdiscovery: extract push-obshashrange discovery in a function This will help us to simplify the code. The code is updated to directly use 'outgoing' missing so that is no longer needs to access 'futureheads'.

obsdiscovery: merge two distinct if statement It did not made any sense for them to be split.

obsdiscovery: extract '_canobshashtree' having this as a simple function will help use to generalise the obsdiscovery algorithm we use.

obsdiscovery: move code inserting discovery at the end of the file This makes it easier to reuser function defined earlier in the file.

obsdiscovery: trust core to do the right thing if no discovery is available This is fixed for a long time now.

obsdiscovery: factor the "do we push markers" check out This clarify the code a bit.

stablerange: also respect the 'warm-cache' config This will help large install to enable obshashrange as "opt-in" without impacting other repositories.

debugstablerange: cleanly "Abort" when no revision are specified Previously we had a traceback from a max on an empty sequence.

stablerange: use last recently used caching for revisions associated to ranges This does not make things faster but at least it make them able to finish on large repositories.

stablerange: report progress more often in slow case For small repositories the time spent in progress was problematic. However on large repository it quickly becomes negligible, so we add a way to detect this case and report more often in this case.

template: update successors template output Use templatekw._hybrid as return value for precursor template, it's nicer and easier to play with than returning a formatted string.

test: extract metaedit tests into test-metaedit.t test file The metaedit command deserver it own test file and this shave of 15% of the runtime of test-evolve.t one of our longest tests.

template: update precursors template output Use templatekw._hybrid as return value for precursor template, it's nicer and easier to play with than returning a formatted string.

template: skip "fate" in test output when empty This should clarify the test output a bit.

template: skip "successors" in test output when empty This should clarify the test output a bit.

template: make the template definition more manageable Readable config is readable

template: skip "Precursors" in test output when empty This should clarify the test output a bit.

template: move each template on its own line during tests This will clarify the various coming changes around template.

refactoring: rename obsfate template to obsfate_quiet As I don't know how to handle templates arguments, create different templates for each verbosity level wanted.

doc: add documentation about effectflag experiment

effectflag: split effect flag meta Split effect flag meta into 3 separate flags: - USERCHANGED - DATECHANGED - BRANCHCHANGED Also display them separately for the moment, they could be grouped back together if we want later.

tests: update missing amend (I need a hook)

template: update obsfate to be more human friendly Approach the obslog output for the moment.

template: add the obsfate template Reuse _getobsoletefate and nextvisiblesuccessors to easily get the obsolescence fate of any node. The limitation is that obsfate is as precise as these two functions, which is not great for nextvisiblesuccessors.

refactor: extract obs fate algorithm from _getobsoletereason Refactor _getobsoletereason to extract the obs fate computation into a separate function, it will be used later and put it in obshistory as it will be needed in other files. Rename _getobsoletereason to _getobsoletefateandsuccessors.

template: add successors template The successors templates display the closest visible successors of each changeset, we may have gap, like in this case: A -> B -> C -> D If we display A, C and D but not B, we can't display than B is the closest successor of A because it's not displayed. We display C as the closest successor of A instead.

template: add precursors template The precursors templates display the closest visible precursors of each changeset, we may have gap, like in this case: A -> B -> C If we display A and C but not B, we can't display than B is the closest precursor of C because it's not displayed. We display A as the closest precursor of C instead. Add a new test file as we will need to generate theses gaps and modifying test-evolve-obshistory.t will lead to many unrelated lines changes.

refactor: extract templates into a new file Extract evolve template from evolve/__init__.py to evolve/templatekw.py as more templates will come in the future.

obshistory: add the all option to obslog to show the while obs tree Add a --all option to obslog in order to display also the successors of each obs markers. It has the effect of showing a more comprehensive graph when a split or a divergence happened.

merge with stable

readme: update changelog for 6.2.2

test: fix test-evolve.t Metaedit take the current date when no date is given. When we added default- date, the date ended up to be the same and the behavior changed. Fix the text by disabling default-date in the metaedit call and update the comment about what is needed to fix correctly the test; take the changeset date when metaediting only one changeset.

test: split evolve cycle tests in a separate test file

evolve: fix --rev with --list This was untested and never worked.

test: improve the globing for obsmarker date If we force the date of obsmarkers, they will no longer include "20".

test: rename all olog references Replace them by hg obslog which is more explicit.

test: improve the globing for obsmarker date If we force the date of obsmarkers, they will no longer include "20".

documentation: recomment using blackbox with the obshashrange experiment

merge with stable

packaging: update the development version number This helps catch non-tagged version installed with pip.

Added tag 6.2.1 for changeset 0af99106b075

packaging: prepare release 6.2.1

compat: merge with stable Stable has multiple changeset that solve deprecation warning for the future Mercurial 4.3.

test-compat: merge with mercurial-3.9

test-compat: merge with mercurial-4.0 branch

test-compat: merge with mercurial-4.1 Nothing to report

test-compat: merge with stable Nothing special to report, minor blackbox output update and color extensions was needed for test-prune.t

deprecation: fix dirstate.beginparentchange deprecation warning Replace calls to dirstart.beginparentchange and dirstate.endparentchange to dirstate.parentchange context manager.

deprecation: fix cmdutil.command deprecation warning

obshistory: be more flexible about createmarkers API The API got touched in core.

inhibit: be more flexible on createmarkers API This API got touched in core.

merge with stable

obshashrange: document the experiment This should help people jump in.

obshashrange: add an option to disable local cache warming The cache still a bit too volatile an costly to update. This offer a way for developer working copy to only compute it when used.

obshashrange: re-warm less revision after reset rewarming every single revision get very expensive quickly. Instead we warm few key revision and rely on the fact this will warm enough useful subrange.

effectflag: simplify meta data change testing Let us just do one larger test. G: changed hgext3rd/evolve/obshistory.py

effectflag: add a small comment to suggest improvement The code run but we will likely need something easier to extend.

effectflag: handle invalid data If we get a marker with an invalid input we should not crash the whole process.

effetflag: move the flag back with the test of the creation logic Previous changeset was published by mistake.

obshistory: display informations from the obs marker effect flag Leverage the effect flag to actually show what change for each obsmarker in the obslog command. (first basic version).

test: add (glob) to globed line Ingloriously forgot to amend.

effectflag: add test combining multiple flag We check that multiple changes might be detected a the same time.

effectflag: basic diff change detection We adds some basic diff detection.

effectflag: detect parent change Add the detection of parent changes.

effectflag: detect meta changes Add the detection of known meta field changes. The known meta fields that we check are: - user - date - branch

effect-flag: rework existing test to prepare for mor tests

effectflag: allow to save effect of change in obsmarkers Obsmarkers lack a precious information, what are the changes between the source and the destination. This could be several thing: - The description has changed, with a amend -e for example - The parent has changed, with a rebase for example - The content has changed, with a amend for example - The meta, like the user, has change, with a amend -u for example Or any combination of them. We hijack the createobsmarkers in order to compute the modification and store the effect flag which is basically a bitfield of the changes list above and we store it in the metadata for the moment. The advantage of computing the flag just before creating the obs marker is that all commands will immediately be compatible and we can start working on the UI display of the flag. As a future optimization, the commands will be able to pass the flag directly or data to help computing them, for example amend knows if the description has changed or rebase now if the parent has changed, virtually removing the overhead. To start simple, this patch add the detection and storage of the effect flag when the description has changed. And hide it behind an experimental flag disabled by default 'experimental.evolution.effect-flags'. More flag will follow.

merge back with stable

readme: fix version number

next: prevent a color related crash when evolve This was not caught by test because they run color less.