search

probonopd / irdb

One of the largest crowd-sourced, manufacturer-independent databases of infrared remote control codes on the web, and aspiring to become the most comprehensive and most accurate one
Other
797 stars 147 forks source link

New files breaking conventions #150

Closed bengtmartensson closed 5 months ago

bengtmartensson commented 5 months ago

Unfortunately, I discovered yesterday that there has been a number of commits recently, that breaks the established conventions for this project.

  1. There has been a number of files that uses tab separated fields, instead the previously used comma separad fields ("csv"). For example: codes/Aiwa/Tapedeck/AD-WX929
  2. Some new remotes do not adhere to the <manufacturer>/<devicetype>/<device>,<subdevice>.csv file name scheme, mandated in the README file. Particularly bad it is when the directory depth is changed, i.e. when codes/*/*/* does not consist of readable files (but possibly directories). For Example: codes/Aiwa/Tapedeck/AD-WX929 (filename do not adhere), codes/Wowee/ Robot Dog/Chippies/Chippies (directory path does not adhere, and ... a leading space in Robot Dog!!)
  3. At least on file contains at least one empty line. codes/RME/ADI DAC2/ADI DAC2

Also for compatibility with many programs, I ask that NEC is not used as protocol name, since it does not contain any repeat informatsion. Instead e.g. NEC1 (or NEC2) should be used, since it repeats.

If this project is to be considered as a data base to be readable from programs, not only humans, these convention violations are really a problem.

Is there a need for discussion (I am more than open for this), or can I just ask for these "hickups" to be fixed?

Some of the new files say "Captured with IrScrutinizer...". Note that said program contains an export format "IRDB CSV" which generates files (but not file names!) in the right format.

@SS72post: Thank you very mucn for your contributions, but they are unfortunately not useful in the present form.

probonopd commented 5 months ago

Ouch. Sorry I didn't catch these.

For the future, maybe we should have some sort of automated test that runs on each pull request and checks the submissions. In the meantime, we should probably roll back the repository to 44744ca?

probonopd commented 5 months ago

Moved what we had so far to jan2024 branch and reset the master branch to before the commits in question.