pypdfium2-team/ctypesgen

Wrapper generator for Python ctypes

ctypesgen's Issues

Ideas

• Improved support for binding to python API.
• Inline-define preamble helpers separately and embed on demand; make c_ptrdiff_t a dependency node and include only as needed
• Add opt-in autostrings helper with configurable codec, see draft
• Merge preamble helpers and loader into a single file (to simplify printer) ? -> problem: headers-only build does not need libraryloader
• Restore multilib support through CLI groups, see draft
• Consider replacing --all-headers with --deepen [N]. Behave like --all-headers if the flag is specified without value, but allow e.g for --deepen 1 to eagerly add members from one nesting only, and not the whole include tree (which might be the most common case to currently use --all-headers).
  Discourage full-depth inclusion (as this tends to spam the bindings and pull in problematic macros) -- instead suggest explicit --symbol-rules yes=....
  (Fun fact: --symbol-rules yes=.+ should lead to the same result as --all-headers)
• Add second CLI entrypoint allowing to bind to each input header individually, as a higher-level layer around current capabilities, i.e. automatic header dependency tree resolution and module linking (within one library). Advantages: namespace separation, closer to original C library.
• Add PYI printer; allow specifying multiple printers at once to avoid repeating common pipeline steps

Fork overview, and thoughts to improve basis for chance of upstreaming

See below for an overview of this fork. Note, this writeup is a non-exhaustive work in progress.

This information may be valuable for working towards a basis that could be merged back into upstream at some point, though this seems fairly hypothetical for the near term, given time constraints, and mismatched design intents (e.g. relating to backwards compatibility).

However, this fork of ctypesgen may be a good starting point for any active future development, with a significantly overhauled code base that should be nicer to work with.

Selection of improvements from this fork

• Removal of bloated old string classes that scream technical debt.
  Enforcement of explicit string encoding/decoding. (We might want to add back implicit string handling as opt-in in the future, see below.)
  Note, the old string classes are incompatible with some python releases of the 3.7/3.8 branches.
  See also pypdfium2-team/pypdfium2#76, ctypesgen#77, python/cpython#16799, ctypesgen#177
• Bloated old library loader replaced with new lean library loader that is more explicit/controllable.
  See also ctypesgen#176, and 569dc4b for some oversights/peculiarities in the old library loader.
  Resolve . to the module directory, not the caller's CWD. Don't add compile libdirs to runtime.
• Preventing the assignment of invalid/non-existent struct fields by correction of __slots__ declaration. This fix should be fairly easy for upstream to pick. See also ctypesgen#183
• Implemented relative imports with --link-modules, and library handle sharing with --no-embed-preamble, Removed incorrect POINTER override. This properly fixes ctypesgen#86 (shared headers), and allows to divide bindings to a library in multiple outputs (e.g. translate each header to a separate python file).
• More powerful/flexible means of control over symbol inclusion via --symbol-rules.
• Pre-processor auto-detection and significant improvements to call style (see 7559e81).
• Removed questionable UNCHECKED wrapper from preamble.
• Do not bypass c_void_p -> int auto-conversion (see Readme or commit for background).
• Propagate exception if no output members were found. (Previously would have been a warning, but the if-check was defunct.)
• New style-related printer options that allow to disable symbol if-guards¹ and macro guards.
• Proper newline concept for the python printer, see a538742.
• Free library handles after use, to allow for in-session deletion of DLLs. This allows to activate a formerly skipped test case on Windows.
• Internal code cleanups and test suite improvements.

small, self-contained fixes have usually been submitted upstream and may have been merged

• Fixed conflicting names resolver never actually being called. 105a3c6, ctypesgen#193. But what the code does is still poor, unfortunately.

Points to consider

• Restoring implicit UTF-8 string encoding/decoding as optional?

  ctypesgen originally did implicit UTF-8 encoding/decoding of in/out strings.
  While that tends to be bad practice and callers had better handle strings explicitly instead, it would seem reasonable to retain an optional backward compatibility layer for existing callers.
  I also imagine it might be convenient for a library that consistently uses UTF-8 for everything.

  Adding the old string classes back is certainly not an option for us. However, it may be possible to create a lean replacement. See ctypesgen#177 for a suggestion (copy below), or ctypesgen#77 (comment) for an alternative draft by @olsonse.
  Note that in/out must be handled in a single class.

• The windows-specific stdcall convention

  Our fork lost it for simplicity while rewriting the library loader. It should be fairly easy to add back, just wondering how to test (as this lies beyond our use case), and how to integrate it nicely.

  Does the calling convention really have to be decided on function level with two library handles for cdecl/stdcall, or would it be sufficient to decide at library level, with a single handle? Is there any example of a single library actually exporting functions with different calling conventions?
  Note that the ctypes API is designed around deciding at library handle level, not at function level, which suggests the expected use case is a library ABI with homogeneous calling convention.

  Possible resolution: Added an option to take a caller-given dll class. It requires a small user interaction and does not support mixed calling conventions, but seems like a nice bloatless way to support a pure stdcall binary.

• Removal of support for multiple libraries in one bindings file

  This feature was a significant complexity burden in some code areas, including pollution around symbols in printer/output code. For now we decided to remove it - callers can use --no-embed-preamble and --link-modules to create separate bindings files. This also encourages individual/explicit rather than unified loader config.

  However, see ctypesgen#86 (comment) for some interesting considerations regarding a possible cleaner re-implementation.

Other notes

• Shifts in design intent: We would prefer to stick with plain ctypes as much as possible and avoid cluttering the bindings with custom wrappers.

• CLI: We changed the command-line interface from action=append to action=extend and nargs=+/*. This implied switching headers from positional to flag argument to avoid confusion/interference with flags that take multiple arguments. There are more CLI changes not listed here, see diff for details.

Done tasks

• Restored test suite usability by adapting to fork changes.
• Restored macro guards as opt-out

Footnotes

1. Note, this is meant for use with inherently ABI correct packaging only ↩

Position-precise imports of linked modules?

Consider passing -dI to the C pre-processor to preserve include statements, in order to later honor the position when translating to an import of a linked module, intending to address the following known limitation (quote from Readme):

Linked modules are naively prioritized in dependency resolver and conflicting names handler, i.e. intentional overrides are ignored. The position of includes is not honored; ctypesgen always imports linked modules at top level.