| Age | Commit message (Collapse) | Author |
|---|
|
5a3c4668 accidentially changed the base directory for cover export
and import files, if given as relative paths. This commit fixes this -
the files are again expected to be given relative to the directory of
the cover spec file itself, or else as absolute paths.
|
|
|
|
Also some code cleanup
|
|
Adjacent separator characters are treated as one in
string:tokens/2. This was shown in the example in the reference manual
but it was not written in words. A note about this is now added.
|
|
|
|
|
|
|
|
We need to make a global budget for testing on CI server and then
define for each property how much it may use of that time. Probably in
auxiliary file, such that it can be computed from what has changed in
the repo.
|
|
We want to be able to see tests with eqc_gen:sample, we also want to be
able to re-run tests.
Side effects are no good idea if these goals need to be met. Therefore,
we replace the side effect to ask for a port by making a symbolic call
of this.
Nicer solutions are possible, but at least we can now re-run the test
case, and therefore shrink.
|
|
The data_dir used by the tests is given at runtime. This has as a
disadvantage that the generate test has a hardcoded data_dir in it
(ssh_eqc_client_server_dirs below):
[{set,{var,1},
{call,ssh_eqc_client_server,initial_state,
[{state,false,[],[],[],[],"ssh_eqc_client_server_dirs"}]}},
{set,{var,2},
{call,ssh_eqc_client_server,ssh_server,
[{{127,1,1,1},
{call,ssh_eqc_client_server,inet_port,[{127,1,1,1}]}},
"ssh_eqc_client_server_dirs",
[{parallel_login,true}]]}},
Re-running this tests on another machine works, since the path is
relative, but if it were absolute, it would have been hard.
Instead, we may use a symbolic representation of the data_dir and fill
it in each time one runs the property, thus even when one does a check
or recheck.
The key to this is to use a variable in the test and bind the variable
in the place where one runs the commands by using the environment
variable feature of run_commands.
Conflicts:
lib/ssh/test/property_test/ssh_eqc_client_server.erl
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* siri/megaco/fix-appup:
[megaco] Fix appup for OTP-17.3
|
|
|
|
|
|
* anders/diameter/17.3_release/OTP-12093:
Add recompilation admonition to 17.2 release notes
|
|
* anders/diameter/Failed-AVP/OTP-12094:
Fix ?MODULE in preprocessed dictionary forms
|
|
That dictionaries need to be recompiled, which is the case whenever
diameter_gen.hrl is modified.
|
|
By replacing literal diameter_gen_relay atoms in forms extracted from
that module by the name of the module in question. This has been wrong
for some time, but only became noticable when the parent commit started
using ?MODULE as more than a process dictionary key or tag to match on.
In particular, the function dict/1 in diameter_gen.hrl (included by
every dictionary module) can now return ?MODULE, which is (not
surprisingly) expected to be the name of the dictionary module in
question. It wasn't in the case of a module compiled from forms: it was
diameter_gen_relay, since that's the module the forms were extracted
from.
The fix only affects dictionaries compiled from forms, as returned by
diameter_make:codec/2. In particular, dictionaries compiled from Erlang
source returned by this function, or by diameterc(1), are unaffected.
|
|
|
|
|
|
|
|
* sverk/dirty-sched-api-cleanup:
erts: Remove enif_have_dirty_schedulers()
erts: Correct dirty scheduler NIF API for Windows
|
|
and add 'dirty_scheduler_support' to ErlNifSysInfo
|
|
|
|
|
|
|
|
The warning is issued if we assign the stream to a variable. The stream
is getting closed together with the socket.
|
|
OtpInputStream and OtoOutputStream don't need closing (they are
ByteArray*Streams)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
was colliding to AbstractNode.flags
|
|
was colliding with OtpLocalNode.port
|
|
field 'OtpLocalNode self' was colliding with 'OtpSelf self' in
OtpConnection
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
