[Panic] index out of bounds

[Radiergummi]

I just experienced a reproducible crash. This seems to be caused by the combination of +D and -D212; it also prints the following warning: 'one-blank-line-before-class' (D203) and 'no-blank-line-before-class' (D211) are incompatible. Ignoring 'one-blank-line-before-class'.

• Command: ruff --select D --ignore D212 src --fix
• Backtrace:

  thread '<unnamed>' panicked at 'index out of bounds: the len is 29 but the index is 29', crates/ruff/src/source_code/locator.rs:83:9
  stack backtrace:
   0:        0x10d897cfa - <std::sys_common::backtrace::_print::DisplayBacktrace as core::fmt::Display>::fmt::h521420ec33f3769d
   1:        0x10d2cc20a - core::fmt::write::h694a0d7c23f57ada
   2:        0x10d894a5c - std::io::stdio::_eprint::hce138c1c2856ce51
   3:        0x10d897aca - std::time::Instant::elapsed::h513af9279b3b413c
   4:        0x10d89bf93 - std::panicking::default_hook::hfeeab2c667b2d7c2
   5:        0x10d89bcf7 - std::panicking::default_hook::hfeeab2c667b2d7c2
   6:        0x10d526270 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
   7:        0x10d89c697 - std::panicking::rust_panic_with_hook::h1b5245192f90251d
   8:        0x10d89c435 - <std::panicking::begin_panic_handler::StrPanicPayload as core::panic::BoxMeUp>::get::h9f4a61dc8b1a8670
   9:        0x10d89a8e8 - <std::sys_common::backtrace::_print::DisplayBacktrace as core::fmt::Display>::fmt::h521420ec33f3769d
  10:        0x10d89c142 - _rust_begin_unwind
  11:        0x10d922713 - core::panicking::panic_fmt::h0097ad8ec0b07517
  12:        0x10d922856 - core::panicking::panic_bounds_check::h1c89633b3be15e5a
  13:        0x10d7036e8 - <ruff::source_code::indexer::Indexer as core::convert::From<&[core::result::Result<(rustpython_compiler_core::location::Location,rustpython_parser::token::Tok,rustpython_compiler_core::location::Location),rustpython_parser::lexer::LexicalError>]>>::from::h495512723e7b0d12
  14:        0x10d703ce8 - ruff::source_code::locator::Locator::slice::h8cda49452a10e3a1
  15:        0x10d64e55c - ruff::linter::lint_fix::h4a770bbd14b0f19a
  16:        0x10d51025a - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  17:        0x10d49ffaa - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  18:        0x10d457931 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  19:        0x10d4a0d67 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  20:        0x10d457a75 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  21:        0x10d4a0d67 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  22:        0x10d457931 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  23:        0x10d4a0d67 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  24:        0x10d457931 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  25:        0x10d4a0d67 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  26:        0x10d4cff25 - <regex_syntax::hir::HirKind as core::fmt::Debug>::fmt::ha120f21aa78097c1
  27:        0x10d92b35a - rayon_core::registry::WorkerThread::wait_until_cold::h38f60dff60493700
  28:        0x10d3f0df7 - rayon_core::registry::ThreadBuilder::run::hd4e0e31ddec10142
  29:        0x10d3ee002 - <quick_xml::events::attributes::AttrError as core::fmt::Debug>::fmt::ha945229a9a8fc49b
  30:        0x10d3ef36e - <quick_xml::events::attributes::AttrError as core::fmt::Debug>::fmt::ha945229a9a8fc49b
  31:        0x10d8a0319 - std::sys::unix::thread::Thread::new::h135242531a117d11
  32:     0x7ff813b49259 - __pthread_start

• ruff.toml:

  select = [
    "E",
    "F",
    "I002",
    "N",
    "UP",
    "PTH",
    "COM",
    "RET",
    "T",
    "S",
    "INP",
    "TCH",
    "SIM",
    "G",
    "ISC",
    "TRY",
    "DJ",
    "ARG",
    "PLW",
    "DTZ",
    "EM",
    "SLF",
  ]
  
  ignore = [
  
    # S101: Assert usage
    # Not using assert is just bogus advice.
    "S101",
  
    # S104: Possible binding to all interfaces
    # We're running in Docker, where binding to all interfaces is explicitly
    # desired behaviour.
    "S104",
  
    # S324: Probable use of insecure hash functions in `hashlib`: `sha1`
    # We don't use SHA1 for confidentiality purposes, so this is okay.
    "S324",
  
    # PLW2901: `for` loop variable `<variable>` overwritten by assignment target
    # When applied judiciously this can make sense.
    "PLW2901",
  
    # TRY003: Avoid specifying long messages outside the exception class
    # Effectively, this suggests to create custom exception classes for every
    # possible error. This makes sense, but doesn't add much value now.
    # TODO: Revisit this decision
    "TRY003",
  
    # D100-D104: Existence of doc comments for public code
    # Again, this makes sense, but doesn't add much value now.
    # TODO: Revisit this decision
    "D100",
    "D101",
    "D102",
    "D103",
    "D104",
  
    # D212: Multi-line docstring summary should start at the first line
    # This just makes the comment harder to read.
    "D212",
  
    # DJ001: Avoid using `null=True` on string-based fields such as CharField
    # We have a reason to make a distinction between "" and NULL, so this is OK.
    "DJ001",
    "DJ008",
  
    # RUF100: Unused `<directive>` directive (non-enabled: `<rule>`)
    # These are usually signs of progressive enhancements and should not crash
    # the commit or build.
    "RUF100",
  
    # SLF001: Private member accessed: `<member>`
    # As we have properties in the index prefixed by a dash, we have to use
    # those names in attrs accessors. Hence, this is not a bug for 99% of cases.
    "SLF001",
  ]
  
  [per-file-ignores]
  # Imports are organized differently in the settings modules. This is important
  # and should not be changed.
  "src/sia/settings/__init__.py" = ["E402", "F403"]
  "src/index/models.py" = ["N815"]
  "src/manage.py" = ["INP001"]
  "src/processing/management/commands/*.py" = ["TRY003", "EM101", "EM102"]
  "src/ai/*.py" = ["T201"]
  "bin/hash_rabbitmq_password.py" = ["INP001"]

[charliermarsh]

Thanks so much! Do you happen to know which file is triggering this? Are you able to include a snippet to reproduce it (or a link to the repo)?

[Radiergummi]

Unfortunately no, I accidentally ran black after this and cannot reproduce the bug anymore :( Feel free to close the issue if this isn't helpful without a repro...

[charliermarsh]

Blerg ok. I'll close for now, but please ping back here if you see this again, I'd love to help!