imphook - Simple and clear import hooks for Python

The imphook module allows to easily define per file type import hooks, i.e. overload or extend import processing for particular file types, without affecting processing of other file types, and at the same time, while ensuring that new processing integrates as seamlessly as possible with normal Python import rules.

Besides the Python-level API to install import hooks, the module also provides command-line interface to run an existing Python script or module with one or more import hooks preloaded (i.e. without modifying existing script source code).

Some but not all things you can easily do using imphook (most of these require additional modules to do the heavy lifting, imphook just allows to plug it seamlessly into the Python import system):

• Override importing of (all or some) .py files, to support new syntax or semantics in them.
• Import files written using a DSL (domain-specific language) as if they were Python modules. E.g., config or data files.
• Import modules written in other language(s), assuming you have an interpreter(s) for them.
• Import binary files, e.g. Java or LLVM bytecode.

imphook works both with new, lightweight legacy-free Python API, as promoted by the Pycopy Python dialect (the original source of the "easy import hooks" idea), and CPython (the older, reference Python implementation), and with other Python implementations which are CPython-compatible.

Quick Start

Make sure that you already installed imphook using:

pip3 install -U imphook

Below is a complete example of an import hook module to load key = value style config files:

import imphook

def hook(modname, filename):
    with open(filename) as f:
        # Create a module object which will be the result of import.
        mod = type(imphook)(modname)
        for l in f:
            k, v = [x.strip() for x in l.split("=", 1)]
            setattr(mod, k, v)
        return mod

imphook.add_import_hook(hook, (".conf",))

Save this as the mod_conf.py file, and add the two following files to test it:

example_settings.conf:

var1 = 123
var2 = hello

example_conf.py:

import example_settings as settings

print(settings.var1)
print(settings.var2)

Now run:

python3 -m imphook -i mod_conf example_conf.py

As you can see, the example_conf.py is able to import example_settings.conf as if it were a normal Python module.

Besides copy-pasting the above and other examples, you can also clone the Git repository of imphook, which contains various ready-to-use examples:

git clone https://github.com/pfalcon/python-imphook

API to install hooks and hook structure

The API of the module consists of one function: imphook.add_import_hook(hook, ext_tuple). hook is a name of hook function. ext_tuple is a tuple of file extensions the hook function should handle (the leading dot should be included). More often than not, you will want to handle just one extension, so don't forget to use the usual Python syntax with a trailing comma for 1-element tuple, e.g.: (".ext",). Python modules may not contain a dot (".") in their names (they are used to separate subpackages), so the extension you register may contain multiple dots, e.g. ".foo.bar", with filename my_module.foo.bar matching it.

It is possible to call imphook.add_import_hook(hook, ext_tuple) multiple times to install multiple hooks. The hooks are installed in the stack-like fashion, the last installed will be called first. It is possible to install multiple hooks for the same file extension, and earlier installed hooks may still be called in this case, because a hook function may skip processing a particular file, and let other hooks to take a chance, with default processing happening if no hook handled the import.

The signature and template of the actual hook function is:

def my_hook(modname, filename):
    # Return None if you don't want to handle `filename`.
    # Otherwise, load `filename`, create a Python module object,
    # with name `modname`, populate it based on the loaded file
    # contents, and return it.

The modname parameter is a full module name of the module to import, in the usual dot-separated notation, e.g. my_module or pkg.subp.mod. For relative imports originated from within a package, this name is already resolved to full absolute name. The modname should be used to create a module object with the given name.

The filename parameter is a full pathname (with extension) of the file which hook should import. This filename is known to exist, so you may proceed to open it directly. You may skip processing this file by returning None from the hook, then other hooks may be tried, and default processing happens otherwise (e.g. .py files are loaded as usual, or ImportError raised for non-standard extensions). For package imports, the value of filename ends with /__init__.py, and that is the way to distinguish module vs package imports.

If the hook proceeds with the file, it should load it by whatever means suitable for the file type. File types which are not natively supported by Python would require installing and using other extension modules (beyond imphook). After loading the file, the hook should create an empty Python module object which will be the result of the import. There are a few ways to do that:

• The baseline is to call a module type as a constructor. To get a module type, just apply the usual type() function to an existing (imported) module. You'll definitely have imphook itself imported, which leads us to:

  mod = type(imphook)(modname)
  

  The parameter to constructor is the name of module to create, as passed to the hook.

• If the above looks too magic for you, you can import symbolic name for module type from the types module:

  from types import ModuleType
  mod = ModuleType(modname)
  

• Finally, you may use the imp module, which may be as well the clearest (to the newbie) way of doing it:

  import imp
  mod = imp.new_module(modname)
  

  But mind that the imp module is considered deprecated.

Of the choices above, the first is the most efficient - no need to import additional modules, and it's just one line. And once you saw and were explained what it does, it shouldn't be a problem to remember and recognize it later.

Once the module object is created as discussed above, you should populate it. A way to do that is by using setattr() builtin to set a particular attribute of a module to a particular value. Attributes usually represent variables with data values, but may be also functions and classes.

Finally, you just return the populated module object.

In case you want to perform custom transformation on the Python source, the process is usually somewhat different, where you transform a representation of the source, and then execute it in the context of a new module, which causes it to be populated. An example of that is provided in the latter section.

Using import hooks in your applications

There are 2 ways to use import hook(s) in you Python programs: either preloading them before starting your program using imphook command-line runner (next section) or load them explicitly at the startup of your application. Crucial thing to remember that import hooks apply: a) for imports only; b) for imports appearing after the hook was installed.

The main file of our application is normally not imported, but executed directly. This leads to the following pattern in structuring your application source files:

• Have a "startup file", which is the one which user will actually run, so name it appropriately. In that file, you load import hooks and perform other baseline system-level initialization.
• The main functionality of your application is contained in seperate module(s). The startup script imports such a main module and executes it (e.g., by calling a function from it).

You already grasped how that works - as the "main" module is imported, whatever hooks the "startup" script installed, will apply to it.

This pattern is actually officially encoded in the structure of Python packages. And any non-trivial Python application will likely be a package with a few sub-modules in it, so you can as well structure your applications this way (as a package), even if they start simple (only one submodule initially). So, if you try to "run" a package, what actually gets run is __main__ submodule in that package. That's exactly the "startup" file we discussed above. It installs the import hooks, and imports a submodule with actual application's functionality.

The actual loading of hooks is very easy: just import them in your startup script, voila! For mod_conf.py hook module shown in the example above that would be:

import mod_conf

You should do that as soon as reasonably possible in your startup file. Normally, that would be after stdlib imports, and before imports of your app's modules. Sometimes, you may want to put hook imports very first, even before the stdlib modules. E.g., if hooks implement JIT compilation, which may benefit even stdlib modules (someone yet has to develop such hooks!). All in all, follow the guidelines above and documentation of the particular hooks that you use.

Finally, the pattern described above (of having "startup" and "main" modules in your app) doesn't work too well in case your application is a single script file, you would need to turn that into 2 files to make the import hooks work. But that's exactly why imphook provides command-line preloader/runner interface!

Command-line interface

Where you would normally run a single script like:

• python3 script.py, or
• python3 -m script

you can run the same script/module with some import hooks preloaded using following commands (changes comparing to the above commands are shown in italics):

• python3 -m imphook -i <mod_hook> script.py, or
• python3 -m imphook -i <mod_hook> -m script

That's exactly how we ran example_conf.py in the Quick Start section. You can repeat -i option multiple times. Alternatively and more compactly, you can pass to single -i option a comma-separated list of hook modules to import, e.g.: -i mod_hook1,mod_hook2,mod_hook3. If you pass multiple hooks, the will be handled in the same stack-like fashion as the API call described above. In the previous example, mod_hook3 will be called first to process imports, then mod_hook2, then mod_hook1. Of course, this will be important only if more than one hook handles the same file extenstion.

This stack-like order on the command-line is used for consistency with the API, to avoid confusion between the two. But it's also the natural order, if you think about it: we start with standard Python import hooks (yes, Python handles all imports using hooks, although its hooks are as simple and clear as those we build here with imphook). Then, there may be some hooks installed in sitecustomize module (that's a way to install some "persistent" hooks for all your projects, which we don't go into, as it should be known for any advanced user). When we get to the imphook command line, we want to be able to override either standard Python or sitecustomize hooks, and that's why all hooks are consistently installed in the stack-like fashion. And you should keep in mind that if an application explicitly installs any hooks, they will have higher priority than those passed on the command line.

We also should again emphasize the difference between script.py and -m script forms above. In the first case, the script is executed directly, and any import hooks you specified with -i do not apply to the script itself (but will apply to imports performed by script.py). Even want hooks to apply to the script's source itself, you must run it using -m script notation (which exactly tells "import this script as a module, don't run it directly"). Pay double attention that when you use -m switch, you must not use the .py extension (if you do, you ask to import the submodule py of package script, which rarely what you want or makes sense).

Last final note is about whitespace in the command-line parameters: they should be used exactly as shown and described: there should always be spaces between -i and -m options and their parameters. And vice versa, if you use comma-separated list of import hooks, there should be no spaces in that list.

Example of Python source transformation

We started this documentation with a quick example of writing an import hook for a simple DSL. We'd like to finish it with examples of another expected common use of imphook - implementing new syntactic (and semantic!) features for Python.

A common starting example is just trying to "rename" one of existing syntactic elements, e.g. use word "function" instead of "lambda".

So, we'd like to get following source code dialect to run:

example_funkw.py:

my_fun = function: print("imphook's functionality is cool!")
my_fun()

The simplest way to do that is just to replace every occurance of "function" in the source with "lambda" (then compile, then execute it in the module context). We thus will come up with the following hook implementation:

mod_funkw_naive.py:

import imphook

def hook(filename):
    with open(filename) as f:
        source = f.read()
    source = source.replace("function", "lambda")
    mod = type(imphook)("")
    exec(source, vars(mod))
    return mod

imphook.add_import_hook(hook, (".py",))

If you read previous sections carefully, you already know that if we want the import hook to apply to the script itself, we must run it as module, using -m switch:

python3 -m imphook -i mod_funkw_naive -m example_funkw

And we get:

imphook's lambdaality is cool!

Oops! The word "lambdaality" is definitely cool, but that's not what we expected! It happens because the code just blindly replaces occurrances everywhere, including within string literals. We could try to work that around by using regular expression replace and match whole words, that would help with the case above, but would still replace lone "function" in the strings. Which makes us conclude: transforming surface representation of a program (i.e. a sequence of characters) is never an adequte method. We should operate on a more suitable program representation, and the baseline such representation is a sequence (or stream) of tokens. Let's use it:

mod_funkw.py:

import tokenize
import imphook

def hook(filename):

    def xform(token_stream):
        for t in token_stream:
            if t[0] == tokenize.NAME and t[1] == "function":
                yield (tokenize.NAME, "lambda") + t[2:]
            else:
                yield t

    with open(filename, "rb") as f:
        # Fairly speaking, tokenizing just to convert back to string form
        # isn't too efficient, but CPython doesn't offer us a way to parse
        # token stream so far, so we have no choice.
        source = tokenize.untokenize(xform(tokenize.tokenize(f.readline)))
    mod = type(imphook)("")
    exec(source, vars(mod))
    return mod

imphook.add_import_hook(hook, (".py",))

Credits and licensing

imphook is (c) Paul Sokolovsky and is released under the MIT license.