Only one instance of base_type() getting a NewRef instead of AdoptRef
fixed in merge. All other changes are superficial formatting and
factoring.
* 'leaks' of https://github.com/MaxKellermann/zeek: (22 commits)
Stmt: use class IntrusivePtr
Stmt: remove unused default constructors and `friend` declarations
Val: remove unimplemented prototype recover_val()
Val: cast_value_to_type() returns IntrusivePtr
Val: use IntrusivePtr in check_and_promote()
Val: use nullptr instead of 0
zeekygen: use class IntrusivePtr
ID: use class IntrusivePtr
Expr: use class IntrusivePtr
Var: copy Location to stack, to fix use-after-free crash bug
Scope: lookup_ID() and install_ID() return IntrusivePtr<ID>
Scope: delete duplicate locals
EventRegistry: automatically delete EventHandlers
main: destroy event_registry after iosource_mgr
zeekygen/IdentifierInfo: delete duplicate fields
main: free the global scope in terminate_bro()
Scope: pop_scope() returns IntrusivePtr<>
Scope: unref all inits in destructor
Var: pass IntrusivePtr to add_global(), add_local() etc.
plugin/ComponentManager: hold a reference to the EnumType
...
Make sure unused scopes are freed to fix memory leaks.
The comment inside pop_scope() is now obsolete and I deleted it,
because this commit implements the real solution.
Note that this requires us to add a reference to the
push_existing_scope() call in dbg_eval_expr(), because it never owned
the reference.
The Zeek code base has very inconsistent #includes. Many sources
included a few headers, and those headers included other headers, and
in the end, nearly everything is included everywhere, so missing
#includes were never noticed. Another side effect was a lot of header
bloat which slows down the build.
First step to fix it: in each source file, its own header should be
included first to verify that each header's includes are correct, and
none is missing.
After adding the missing #includes, I replaced lots of #includes
inside headers with class forward declarations. In most headers,
object pointers are never referenced, so declaring the function
prototypes with forward-declared classes is just fine.
This patch speeds up the build by 19%, because each compilation unit
gets smaller. Here are the "time" numbers for a fresh build (with a
warm page cache but without ccache):
Before this patch:
3144.94user 161.63system 3:02.87elapsed 1808%CPU (0avgtext+0avgdata 2168608maxresident)k
760inputs+12008400outputs (1511major+57747204minor)pagefaults 0swaps
After this patch:
2565.17user 141.83system 2:25.46elapsed 1860%CPU (0avgtext+0avgdata 1489076maxresident)k
72576inputs+9130920outputs (1667major+49400430minor)pagefaults 0swaps
This commit marks (hopefully) ever one-parameter constructor as explicit.
It also uses override in (hopefully) all circumstances where a virtual
method is overridden.
There are a very few other minor changes - most of them were necessary
to get everything to compile (like one additional constructor). In one
case I changed an implicit operation to an explicit string conversion -
I think the automatically chosen conversion was much more convoluted.
This took longer than I want to admit but not as long as I feared :)
This follows rather closely how rotation currently works in
rotate-logs.bro. logging.bro now defines:
# Default rotation interval; zero disables rotation.
const default_rotation_interval = 0secs &redef;
# Default naming suffix format.
const default_rotation_date_format = "%y-%m-%d_%H.%M.%S" &redef;
# Default postprocessor for writers outputting into files.
const default_rotation_postprocessor = "" &redef;
# Default function to construct the name of the rotated file.
# The default implementation includes
# default_rotation_date_format into the file name.
global default_rotation_path_func: function(info: RotationInfo) : string &redef;
Writer support for rotation is optional, usually it will only make
sense for file-based writers.
TODO: Currently, there's no way to customize rotation on a per file
basis, there are only the global defaults as described above.
Individual customization is coming next.
* origin/topic/gregor/bif-tuning:
Refactor: BifTypePtr --> BifType
Bif const: make sure const is indeed a constant.
Support any type in bif const declaration.
Tweak for bifcl
Fix to bifcl wrt namespaces.
Enable declaration of set, vector, and table types in bifs.
Moving type declarations into its own bif file
Support namespaces / modules in bif. Checkpoint.
Support namespaces / modules in bif. Checkpoint.
Remove leftovers from removing "declare enum" from bifcl
Use namespaces for NetVar type pointers.
Remove unused and unnecessary "declare enum" from bifcl
Bif: add record type declaration.
Minor tweaks for bif language.
enum type: don't allow mixing of explicit value and auto-increment.
Add support for enum with explicit enumerator values.
Closes#403.
(now actually commiting all the files)
This change is actually two-fold:
a) bif's now accept module XYZ; statements and module::ID for
function, const, event, enum, etc. declartation
b) Added C++-namespaces to variables, functions, etc. that are declared
in bif but accessed from C++
This required some (lightweight) re-factoring of the C++ codes.
Note, event's don't have their own C++ namespace yet, since this
would require a rather huge re-factoring.
Compiles and passes test suite.
New namespace feature not tested yet.
Documentation to follow.
* origin/topic/seth/fix-compiler-warnings:
Fixed problem with PRI macros.
PRI macros are currently not working for some reason.
Two more small compile time error fixes.
Cleaned up the output from running binpac.
Added line to expect shift/reduce errors in parse.in
Cleaned up g++ warnings.
Addition: I fixed a few more warnings I was getting, and tweaked some
of the existing changes slightly.
