nsf

Clone Tools
  • last updated 2 hours ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
- better to let file classify this file as latex than C++

- removed c++ mode

- fix typo

nsf.c: - Reform of subclass/superclass handling. * unifying transitive subclass handling * localizing and documenting caching of subclass/superclass lists * eliminating repetitious computation of subclass lists * refactored code, renamed some functions to better refelect their purpose * improved documentation - fixed a potential crash for class deletion triggering implicit deletions further deletions referencing the parent class - extended regression test

- fix spelling

all artifacts saved

- improve wording of comment

- Avoid duplicate entries in instance lists for diamond inheritance by maintaining the set of already processed entries in NsfClassInfoInstancesMethod1() and GetAllInstances(). - extended regression test - removed "namespace import" in objectsystem test

Merge branch 'master' of ssh://alice.wu-wien.ac.at/usr/local/src/git-repo/nsf

- handle duplicates in the cmd-list during cleanup

- generic/nsf.c: DispatchUnknownMethod() In the re-entrant case, the unknown dispatcher was too greedy by skipping the first element in the objv at basically all times. For CallConfigureMethod() case in XOTcl2, the actual method name was so skipped. Generally speaking, the "skipping rules" in DispatchUnknownMethod() feel a touch too particular to me: For instance, the dispatch flag "-local", "-system" etc. must be handled. Also, unknowns in method ensembles have the unknown method listed first in objv. Alternatively, we might review how objc/objv make it into DispatchUnknownMethod() (there are just two call sites). But this would merely shift/spread the skipping. Changing the arg vectors, to begin with, (e.g. removing "-local", "-system", etc. from the objv) would requiring rebuilding the arg vector. - tests/parameters.test: Added a simplistic test case to catch the error reporting for XOTcl 2's residualargs handling.

- close parens in Annouce file

- updated ChangeLog

- adding anouncement for the release of 1.6.7

- bumped version numner to 1.6.7

- fixed 2 more version numbers for xotcl 1.6.7

* xotcl.c: some c-cleanup * library files: require exact version numbers to avoid conflicts with XOTcl 2.0

- nsf.c: fix call of DispatchDefaultMethod() in cases it triggers an error for 8.6b - nx.tcl: give a slightly better error message in case the root-object of an ensemble is called directly

- updated TODO

nsf.c: - added method NsfMethodNamePath() to compute the ensemble path in error messages - reduce verbosity - extended regression test

- minor reformatting

nsf.c: - removed conditionals in AddInstance() and RemoveInstance() - dropped result from RemoveInstance()

- Replaced hash-table for GetAllInstances() with a linear list. As a result, mass-destroy on exit is now much faster. Valgrind reports that the full cicrcle of creating 100.000 objects and destroying it on exit became about 15% faster. - added additional argument adEnd to CmdListAdd() - renamed CmdListRemoveList() to CmdListFree() - improved code documentation - removed unneeded AddInstance() and RemoveInstance() - deactivated uncalled ReverseLookupCmdFromCmdTable() and GetHiddenObjectFromCmd() since apparently not needed any more

- fixed documentation bug in migration guide, since arguments in "info slot ..." were deleted in the code, but not in the guide

- fix requiredness of subcmd in "info method subcmd"

- bump version number to 2.0b3

- library/nx/nx.tcl: Some slight renamings for the sake of consistency in the docs

- library/nx/nx.nxd: Rewrote and completed API doc for nx::Class

- library/lib/nxdoc-assets/submethod.html.yuidoc: Supporting origin resolution for submethods.

- TODO: Noted down some observations, issues. To be reviewed. - library/nx/nx.nxd: Completed doc of ::nx::Object - library/nx/nx.tcl: While documenting ::nx::Object, I applied some changes along the way. For example, adding a missing element to the require() API, removing some unwanted params from the "info slot *" sub-interface, added some parameter types where appropriate, ...