commit d6f93b022d
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 17:36:52 2021 +0100
add tests which validate the /status/latest call.
The URL which returns the latest status for all the devices at once is
validated. The tests are not extensive. They just check to make sure
all the intended device are present and the last device on the list is
'device 6'. If any of the device names change or 'device 6' is not the
last device, it means the API has changed.
commit 75d4748150
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 16:51:59 2021 +0100
rename light meter status tests (factory devices).
Rename the tests to make it easier to read and make the naming
convention align with the status tests for the gallery devices.
commit 4d8a1e6f9b
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 16:43:47 2021 +0100
correct 'reading' tests to 'status' tests and add new tests.
I managed to not notice I wrote tests which check the status (on/off)
of the devices but labelled the test as 'reading data' checks. I've
relabelled the tests to avoid the confusion.
I've added additional test which validate the data produced by the
devices in the gallery (part of the Return to Ritherdon project). And,
I've written tests which actually validate the light reading data
produced by the devices in the welding booths in Ritherdon (again,
part of the Return to Ritherdon project).
The new tests include code from the Ratify package. I've only used the
datetime-p function for now. It checks if the timestamps produced are
valid datetime strings -- and can be parsed as such.
commit c9559f0a67
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 16:41:18 2021 +0100
add Ratify to the testing system.
At the moment, this is test the timestamps returned from the API. If
Ratify can parse the timestamps, it returns them. If it can't parse
them, it returns NIL.
commit 2ebd6d2a0f
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 00:45:36 2021 +0100
writing tests for validating light reading data.
These tests check the shape of the data produced by the light meters
in the welding in booths in Ritherdon. I've place a note with the
factory3 tests explaining the reason why there is a factory3 in the
system but there is not actual light meter in the third welding booth
in Ritherdon.
commit 77952ad294
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 00:40:35 2021 +0100
ASDF package definition changes.
The changes here are done out of frustration and desperation. SLIME
was in a right mess (caching issue I assume) so I started changing
things in here to try and fix it. This was before I stumbled on the
'restart Emacs/SLIIME' solution. Becuase it's now working, I've just
kept the changes in place. The relief of getting 'asdf:test-system'
working after about two hours of faffing was enough for me to not
touch this file again. I might come back to this at a later date --
see if/how it needs cleaning up. I might just skip it and start a new
project -- this is a learning exercise after all.
commit d11119bfdc
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 00:37:31 2021 +0100
export additional test function (at package level).
This is so you can run the tests with ASDF, I think. It was something
I did whilst SLIME was a bit messes-up and I was a bit desperate with
trying to fix it. So, I don't know if this additional exported
function is needed. Everything seems to work with it still there --
after I restarted Emacs and SLIME -- so I'm keeping it like this for now.
commit b055b97e2a
Author: Craig Oates <craig@craigoates.net>
Date: Fri Jul 2 00:32:52 2021 +0100
rename tests file to main.lisp.
This was not needed in the end. I was having trouble with SLIME and I
was getting desperate at one point and starting changing
everything. It turns out I just needed to restart SLIME. At the time
of writing, I didn't know this or how to do that so I only noticed
when I restarted Emacs. It looks like stuff got messed-up in the cache
and SLIME couldn't find files/packages because it was looking for old
one which were either renames or deleted (due to me frantically
changing things).
commit 8c87cf5b37
Author: Craig Oates <craig@craigoates.net>
Date: Wed Jun 30 01:44:45 2021 +0100
fix typo in .gitignore (.directory)
Git is picking up my systems .directory files. They are not part of
this project's code base. I accidently put /.directory in a previous
commit instead of .directory.
commit 09dd5f9b6e
Author: Craig Oates <craig@craigoates.net>
Date: Wed Jun 30 01:40:49 2021 +0100
add functions and comments for making HTTP-Requests.
These are comments and functions to get me into the swing of writting
Common Lisp. The code here is the main part of the system but this
commit is mostly about how this file and its code fits into the bigger
picture which is the system/code base.
commit e912fdc73d
Author: Craig Oates <craig@craigoates.net>
Date: Wed Jun 30 01:38:36 2021 +0100
add drakma and cl-json packages to project.
These packages are for making an HTTP-Request (to Ritherdon REST API)
and parse the JSON returned.
commit 05b876bea7
Author: Craig Oates <craig@craigoates.net>
Date: Tue Jun 29 21:38:48 2021 +0100
add 'hello' function (for quick test after loading package).
This is so I can make sure I know ritherdon-rest.lisp can/is loaded
correctly -- via Quicklisp or ASDF -- and is ready to start adding the
actual code.
commit 8f8cd147ce
Author: Craig Oates <craig@craigoates.net>
Date: Tue Jun 29 21:34:14 2021 +0100
integrate FiveAM into test package def. and main testing file.
This builds on the initial set-up in the .asd file. With the .asd file
knowing the tests package needs FiveAM, the code here integrates the
testing framework in to the .lisp files responsible for housing the
tests.
The code here is placeholder tests and should be deleted the more I
get into the project. They exist just to make sure everything is
set-up properly between the various definition/set-up files.
commit 943a912d96
Author: Craig Oates <craig@craigoates.net>
Date: Tue Jun 29 21:32:11 2021 +0100
connect fiveAM testing package to main project in .asd file.
This is so you can use asdf:test-system by just calling the
ritherdon-rest project -- making it easier to work with.
commit 26b025361d
Author: Craig Oates <craig@craigoates.net>
Date: Tue Jun 29 21:31:11 2021 +0100
remove doc folder.
This is beyond where I'm at right now so going to leave it for a
future project.
commit aa1abec7fe
Author: Craig Oates <craig@craigoates.net>
Date: Tue Jun 29 00:22:44 2021 +0100
add fiveAM package and create initial test.
I've changed how the tests and doc systems are defined in the .asd
file. The changes are based on what SLIME outputted when compiling the
project.
The initial test are irrelevant to the project. I wrote it to make
sure fiveAM (and the test project as a whole) was connected together
properly. This test will (should) not remain once the main code is up
and running.
commit 00ba38f3b0
Author: Craig Oates <craig@craigoates.net>
Date: Mon Jun 28 23:11:46 2021 +0100
create project files.
The URL which returns the latest status for all the devices at once is
validated. The tests are not extensive. They just check to make sure
all the intended device are present and the last device on the list is
'device 6'. If any of the device names change or 'device 6' is not the
last device, it means the API has changed.
I managed to not notice I wrote tests which check the status (on/off)
of the devices but labelled the test as 'reading data' checks. I've
relabelled the tests to avoid the confusion.
I've added additional test which validate the data produced by the
devices in the gallery (part of the Return to Ritherdon project). And,
I've written tests which actually validate the light reading data
produced by the devices in the welding booths in Ritherdon (again,
part of the Return to Ritherdon project).
The new tests include code from the Ratify package. I've only used the
datetime-p function for now. It checks if the timestamps produced are
valid datetime strings -- and can be parsed as such.
At the moment, this is test the timestamps returned from the API. If
Ratify can parse the timestamps, it returns them. If it can't parse
them, it returns NIL.
These tests check the shape of the data produced by the light meters
in the welding in booths in Ritherdon. I've place a note with the
factory3 tests explaining the reason why there is a factory3 in the
system but there is not actual light meter in the third welding booth
in Ritherdon.
This is so you can run the tests with ASDF, I think. It was something
I did whilst SLIME was a bit messes-up and I was a bit desperate with
trying to fix it. So, I don't know if this additional exported
function is needed. Everything seems to work with it still there --
after I restarted Emacs and SLIME -- so I'm keeping it like this for now.
This was not needed in the end. I was having trouble with SLIME and I
was getting desperate at one point and starting changing
everything. It turns out I just needed to restart SLIME. At the time
of writing, I didn't know this or how to do that so I only noticed
when I restarted Emacs. It looks like stuff got messed-up in the cache
and SLIME couldn't find files/packages because it was looking for old
one which were either renames or deleted (due to me frantically
changing things).
This builds on the initial set-up in the .asd file. With the .asd file
knowing the tests package needs FiveAM, the code here integrates the
testing framework in to the .lisp files responsible for housing the
tests.
The code here is placeholder tests and should be deleted the more I
get into the project. They exist just to make sure everything is
set-up properly between the various definition/set-up files.
I've changed how the tests and doc systems are defined in the .asd
file. The changes are based on what SLIME outputted when compiling the
project.
The initial test are irrelevant to the project. I wrote it to make
sure fiveAM (and the test project as a whole) was connected together
properly. This test will (should) not remain once the main code is up
and running.
Craig Oates