Revert "enable cpplint"

[sshane]

Reverts #2238

Breaks openpilot compilation

batman@workstation-shane:~/openpilot$ scons -j32
scons: Reading SConscript files ...
scons: done reading SConscript files.
scons: Building targets ...
Retrieved `opendbc_repo/opendbc/can/common.os' from cache
Retrieved `opendbc_repo/opendbc/can/dbc.os' from cache
Retrieved `opendbc_repo/opendbc/can/parser.os' from cache
Retrieved `opendbc_repo/opendbc/can/packer.os' from cache
Retrieved `opendbc_repo/opendbc/safety/tests/libsafety/safety.os' from cache
Retrieved `opendbc_repo/opendbc/can/libdbc.so' from cache
Retrieved `opendbc_repo/opendbc/safety/tests/libsafety/libsafety.so' from cache
arm-none-eabi-gcc -o panda/board/jungle/main-panda_jungle.o -c -mcpu=cortex-m4 -mhard-float -DSTM32F4 -DSTM32F413xx -Iboard/stm32f4/inc -mfpu=fpv4-sp-d16 -fsingle-precision-constant -Os -g -DPANDA_JUNGLE -DALLOW_DEBUG -Wall -Wextra -Wstrict-prototypes -Werror -mlittle-endian -mthumb -nostdlib -fno-builtin -std=gnu11 -fmax-errors=1 -Tpanda/board/stm32f4/stm32f4_flash.ld -Ipanda/board/jungle -Ipanda/board -Ipanda -Ipanda/board -Iopendbc/safety panda/board/jungle/main.c
Retrieved `opendbc_repo/opendbc/can/packer_pyx.so' from cache
Retrieved `opendbc_repo/opendbc/can/parser_pyx.so' from cache
arm-none-eabi-gcc -o panda/board/jungle/main-panda_jungle_h7.o -c -mcpu=cortex-m7 -mhard-float -DSTM32H7 -DSTM32H725xx -Iboard/stm32h7/inc -mfpu=fpv5-d16 -fsingle-precision-constant -Os -g -DPANDA_JUNGLE -DALLOW_DEBUG -Wall -Wextra -Wstrict-prototypes -Werror -mlittle-endian -mthumb -nostdlib -fno-builtin -std=gnu11 -fmax-errors=1 -Tpanda/board/stm32h7/stm32h7x5_flash.ld -Ipanda/board/jungle -Ipanda/board -Ipanda -Ipanda/board -Iopendbc/safety panda/board/jungle/main.c
In file included from panda/board/jungle/main.c:4:
opendbc/safety/safety.h:3:10: fatal error: safety/safety_declarations.h: No such file or directory
    3 | #include "safety/safety_declarations.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
scons: *** [panda/board/jungle/main-panda_jungle.o] Error 1
arm-none-eabi-gcc -o panda/board/main-panda.o -c -mcpu=cortex-m4 -mhard-float -DSTM32F4 -DSTM32F413xx -Iboard/stm32f4/inc -mfpu=fpv4-sp-d16 -fsingle-precision-constant -Os -g -DPANDA -DALLOW_DEBUG -Wall -Wextra -Wstrict-prototypes -Werror -mlittle-endian -mthumb -nostdlib -fno-builtin -std=gnu11 -fmax-errors=1 -Tpanda/board/stm32f4/stm32f4_flash.ld -Ipanda/board -Ipanda -Ipanda -Ipanda/board -Iopendbc/safety panda/board/main.c
In file included from panda/board/jungle/main.c:4:
opendbc/safety/safety.h:3:10: fatal error: safety/safety_declarations.h: No such file or directory
    3 | #include "safety/safety_declarations.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
scons: *** [panda/board/jungle/main-panda_jungle_h7.o] Error 1
In file included from panda/board/main.c:13:
opendbc/safety/safety.h:3:10: fatal error: safety/safety_declarations.h: No such file or directory
    3 | #include "safety/safety_declarations.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
scons: *** [panda/board/main-panda.o] Error 1
scons: building terminated because of errors.

[adeebshihadeh]

How can we be more robust to this?

[aubsw]

Sorry about the breakage.

Since OP currently builds with opendbc@master, there are two options that I can think of:

A. peg opendbc to a specific version in OP repo
B. add a premerge step to opendbc PRs to check that OP compiles with new change

I assume you don't want to do A. for sake of development velocity, but let me know if you think otherwise.

[sshane]

I think the port from panda to opendbc was a bit hacky to get the paths working properly, we can probably refactor that a ton

[aubsw]

we can probably refactor that a ton

Ahaha this is what I'm trying to do in commaai/panda#2171 (comment) 🫠

The main problem is that there's duplicated code in opendbc and panda, which makes it extremely are to refactor without compilation errors AND without breaking OP.

If you don't mind breaking OP for a few minutes, I have 3 PRs that will fix the duplicated code problem. This will in turn allow us to land the cpplint PR.

They need to be merged in this order:

1. Forward-declare dlc_to_len #2248
2. Delete duplicated code from panda panda#2186
3. Get dlc_to_len from opendbc openpilot#35239

Please let me know if you see a different path... This is all I have after massaging this for a really long time.

[sshane]

Ideally we can do this piece wise without putting opendbc in a bad state for external projects. Revert one and suddenly something is broken

[aubsw]

Just to be clear, the changes proposed above never put opendbc itself in a bad state–it does temporarily break panda and OP tho.

We have 3 options:

1. keep Forward-declare dlc_to_len #2248 as-is
2. Make dlc_to_len not static
3. Revive can.h but still move the definition of dlc_to_len

Do you have a preference?

[aubsw]

I thought about it some more. Option #2 is the least problematic option. We can rely on MISRA checks to prevent divergent declarations of dlc_to_len.

I updated the PR accordingly.