Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

register teal.logger handlers #628

Merged
merged 5 commits into from
Mar 20, 2024
Merged

register teal.logger handlers #628

merged 5 commits into from
Mar 20, 2024

Conversation

pawelru
Copy link
Contributor

@pawelru pawelru commented Feb 1, 2024

close https://github.com/insightsengineering/coredev-tasks/issues/502
test with insightsengineering/teal.logger#73

As a bonus, we can safely move logger to suggests (it's still used in tests)

@pawelru pawelru marked this pull request as ready for review March 11, 2024 16:55
Copy link
Contributor

github-actions bot commented Mar 11, 2024

Unit Tests Summary

 1 files   6 suites   0s ⏱️
29 tests 29 ✅ 0 💤 0 ❌
67 runs  67 ✅ 0 💤 0 ❌

Results for commit 657f3dd.

♻️ This comment has been updated with latest results.

@pawelru pawelru added the core label Mar 14, 2024
R/tm_a_pca.R Show resolved Hide resolved
Copy link
Contributor

@gogonzo gogonzo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tusind tak!

pawelru added a commit to insightsengineering/teal.logger that referenced this pull request Mar 20, 2024
close insightsengineering/coredev-tasks#502
#69

I tried to use `logger::log_messages()` (and similar functions for
warnings and errors) but I have encounter following problems:
- this is logging into the global logger namespace whereas we want
particular (package) namespace instead - created an issue
- I cloned logger locally and fixed / enhanced the above but then I
realized that the (global) message / warning / stop calls from the
outside of a package will also log to that package logger namespace. In
other words, these are _global_ handlers for all the warnings and not
only warnings created inside `teal` e.g.
```
r$> warning(1)
[WARN] 2024-01-26 15:58:17.0061 pid:83248 token:[] teal 1
Warning message:
1
```

Therefore I have decided to implement own register handlers mechanism. I
initially started doing this separately inside each package but then I
realised that it's not so DRY and decided to extend core `teal.logger`
functionality.

This is not a reprex as I modified `teal` and `teal.slice` namespaces in
a following way:
- added `register_handlers(<pkg>)` in both packages
- For debugging: added following definitions:
  - `teal.slice`:
    - `teal_s_message` calling `message()`
    - `teal_s_warning` calling `warning()`
    - `teal_s_stop` calling `stop()`
  - in `teal`:
    - `teal_message` calling `message()`
    - `teal_warning` calling `warning()`
    - `teal_stop` calling `stop()`
- `teal_message_nested` calling `teal_s_message()` (so as to simulate
`teal` calling `teal.slice`)
    - `teal_warning_nested` calling `teal_s_warning()`
    - `teal_stop_nested` calling `teal_s_stop()`

The outcome:

```
r$> devtools::load_all("teal.logger")
r$> devtools::load_all("teal.slice")
r$> devtools::load_all("teal")

r$> message(1)
1
r$> teal_message(1)
[INFO] 2024-01-26 15:57:40.8901 pid:83248 token:[] teal 1
1
r$> teal_s_message(1)
[INFO] 2024-01-26 15:57:42.1464 pid:83248 token:[] teal.slice 1
1
r$> teal_message_nested(1)
[INFO] 2024-01-26 15:57:50.9231 pid:83248 token:[] teal 1
[INFO] 2024-01-26 15:57:50.9257 pid:83248 token:[] teal.slice 1
1

r$> warning(2)
Warning message:
2 
r$> teal_s_warning(2)
[WARN] 2024-01-26 15:58:17.0061 pid:83248 token:[] teal.slice In ‘teal_s_warning(2)’: 2
Warning message:
In teal_s_warning(2) : 2
r$> teal_warning(2)
[WARN] 2024-01-26 15:58:18.4165 pid:83248 token:[] teal In ‘teal_warning(2)’: 2
Warning message:
In teal_warning(2) : 2
r$> teal_warning_nested(2)
[WARN] 2024-01-26 15:58:20.8666 pid:83248 token:[] teal In ‘teal.slice:::teal_s_warning(message)’: 2
[WARN] 2024-01-26 15:58:20.8685 pid:83248 token:[] teal.slice In ‘teal.slice:::teal_s_warning(message)’: 2
Warning message:
In teal.slice:::teal_s_warning(message) : 2

r$> stop(3)
Error: 3
r$> teal_s_stop(3)
[ERROR] 2024-01-26 15:58:32.9558 pid:83248 token:[] teal.slice In ‘teal_s_stop(3)’: 3
Error in teal_s_stop(3) : 3
r$> teal_stop(3)
[ERROR] 2024-01-26 15:58:33.5246 pid:83248 token:[] teal In ‘teal_stop(3)’: 3
Error in teal_stop(3) : 3
r$> teal_stop_nested(3)
[ERROR] 2024-01-26 15:58:34.0213 pid:83248 token:[] teal In ‘teal.slice:::teal_s_stop(message)’: 3
[ERROR] 2024-01-26 15:58:34.0226 pid:83248 token:[] teal.slice In ‘teal.slice:::teal_s_stop(message)’: 3
Error in teal.slice:::teal_s_stop(message) : 3
```

Interpretation (where "m/w/s" stands for "message / warning / stop"):
- global m/w/s is not impacted - we only capture m/w/s called from
inside the package
- capturing m/w/s in the appropriate logger namespace - see it as a part
of logger produced string - second last part
- the case of nested calls looks odd at the very first glance but I
think it's correct. Each (package) logger instance might have its own
log threshold, appender function, layout etc. It's expected to have
duplicated logs for each of the registered package namespace found on
the stack of calls. I'm interested in hearing your thoughts on this


PRs:

- [ ] insightsengineering/teal#1081
- [ ] insightsengineering/teal.goshawk#258
- [ ]
insightsengineering/teal.modules.clinical#1010
- [ ]
insightsengineering/teal.modules.general#628
- [ ]
insightsengineering/teal.modules.hermes#354
- [ ] insightsengineering/teal.osprey#252
- [ ] insightsengineering/teal.slice#551
- [ ] insightsengineering/teal.transform#175

(I'll keep them as draft as this is conflicting with release plans
because teal.logger would have to be released first)

TODO: tests - however I'm not yet sure how to make them "clean"

---------

Signed-off-by: Pawel Rucki <[email protected]>
Co-authored-by: Dawid Kałędkowski <[email protected]>
DESCRIPTION Outdated Show resolved Hide resolved
Copy link
Contributor

badge

Code Coverage Summary

Filename                      Stmts    Miss  Cover    Missing
--------------------------  -------  ------  -------  ------------------------------------
R/tm_a_pca.R                    827     827  0.00%    108-1068
R/tm_a_regression.R             776     776  0.00%    153-1033
R/tm_data_table.R               179     179  0.00%    93-325
R/tm_file_viewer.R              170     170  0.00%    44-248
R/tm_front_page.R               129     118  8.53%    70-222
R/tm_g_association.R            334     334  0.00%    135-541
R/tm_g_bivariate.R              671     411  38.75%   306-771, 812, 923, 940, 958, 969-991
R/tm_g_distribution.R          1042    1042  0.00%    122-1299
R/tm_g_response.R               350     350  0.00%    154-577
R/tm_g_scatterplot.R            726     726  0.00%    230-1057
R/tm_g_scatterplotmatrix.R      282     263  6.74%    165-476, 537, 551
R/tm_missing_data.R            1058    1058  0.00%    92-1304
R/tm_outliers.R                 988     988  0.00%    134-1265
R/tm_t_crosstable.R             255     255  0.00%    141-444
R/tm_variable_browser.R         825     820  0.61%    79-1064, 1102-1286
R/utils.R                        99      96  3.03%    82-267
R/zzz.R                           2       2  0.00%    2-3
TOTAL                          8713    8415  3.42%

Diff against main

Filename      Stmts    Miss  Cover
----------  -------  ------  --------
R/zzz.R          +1      +1  +100.00%
TOTAL            +1      +1  -0.00%

Results for commit: 657f3dd

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

@pawelru pawelru enabled auto-merge (squash) March 20, 2024 14:39
@pawelru pawelru merged commit 2892295 into main Mar 20, 2024
24 checks passed
@pawelru pawelru deleted the register_handlers branch March 20, 2024 14:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants