contrib/fuzz/mpatch.cc

packaging: use PyOxidizer for producing WiX MSI installer

We recently taught our in-tree PyOxidizer configuration file to
produce MSI installers with WiX using PyOxidizer's built-in
support for doing so.

This commit changes our WiX + PyOxidizer installer generation
code to use this functionality.

After this change, all the Python packaging code is doing is the
following:

* Building HTML documentation
* Making gettext available to the build process.
* Munging CLI arguments to variables for the `pyoxidizer` execution.
* Invoking `pyoxidizer build`.
* Copying the produced `.msi` to the `dist/` directory.

Applying this stack on stable and rebuilding the 5.8 MSI installer
produced the following differences from the official 5.8 installer:

* .exe and .pyd files aren't byte identical (this is expected).
* Various .dist-info/ directories have different names due to older
  versions of PyOxidizer being buggy and not properly normalizing
  package names. (The new behavior is correct.)
* Various *.dist-info/RECORD files are different due to content
  divergence of files (this is expected).
* The python38.dll differs due to newer PyOxidizer shipping a newer
  version of Python 3.8.
* We now ship python3.dll because PyOxidizer now includes this file
  by default.
* The vcruntime140.dll differs because newer PyOxidizer installs a newer
  version. We also now ship a vcruntime140_1.dll because newer versions
  of the redistributable ship 2 files now.

The WiX GUIDs and IDs of installed files have likely changed as a
result of PyOxidizer's different mechanism for generating those
identifiers. This means that an upgrade install of the MSI will
replace files instead of doing an incremental update. This is
likely harmless and we've incurred this kind of breakage before.

As far as I can tell, the new PyOxidizer-built MSI is functionally
equivalent to the old method. Once we drop support for Python 2.7
MSI installers, we can delete the WiX code from the repository.

This commit temporarily drops support for extra `.wxs` files. We
raise an exception instead of silently not using them, which I think
is appropriate. We should be able to add support back in by injecting
state into pyoxidizer.bzl via `--var`. I just didn't want to expend
cognitive load to think about the solution as part of this series.

Differential Revision: https://phab.mercurial-scm.org/D10688

/*
 * mpatch.cc - fuzzer harness for mpatch.c
 *
 * Copyright 2018, Google Inc.
 *
 * This software may be used and distributed according to the terms of
 * the GNU General Public License, incorporated herein by reference.
 */
#include <iostream>
#include <memory>
#include <stdint.h>
#include <stdlib.h>
#include <vector>

#include "fuzzutil.h"

extern "C" int LLVMFuzzerInitialize(int *argc, char ***argv)
{
	return 0;
}

// To avoid having too many OOMs from the fuzzer infrastructure, we'll
// skip patch application if the resulting fulltext would be bigger
// than 10MiB.
#define MAX_OUTPUT_SIZE 10485760

extern "C" {
#include "bitmanipulation.h"
#include "mpatch.h"

struct mpatchbin {
	std::unique_ptr<char[]> data;
	size_t len;
};

static mpatch_flist *getitem(void *vbins, ssize_t pos)
{
	std::vector<mpatchbin> *bins = (std::vector<mpatchbin> *)vbins;
	const mpatchbin &bin = bins->at(pos + 1);
	struct mpatch_flist *res;
	LOG(2) << "mpatch_decode " << bin.len << std::endl;
	if (mpatch_decode(bin.data.get(), bin.len, &res) < 0)
		return NULL;
	return res;
}

// input format:
// u8 number of inputs
// one u16 for each input, its length
// the inputs
int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size)
{
	if (!Size) {
		return 0;
	}
	// First byte of data is how many texts we expect, first text
	// being the base the rest being the deltas.
	ssize_t numtexts = Data[0];
	if (numtexts < 2) {
		// No point if we don't have at least a base text and a delta...
		return 0;
	}
	// Each text will be described by a byte for how long it
	// should be, so give up if we don't have enough.
	if ((Size - 1) < (numtexts * 2)) {
		return 0;
	}
	size_t consumed = 1 + (numtexts * 2);
	LOG(2) << "input contains " << Size << std::endl;
	LOG(2) << numtexts << " texts, consuming " << consumed << std::endl;
	std::vector<mpatchbin> bins;
	bins.reserve(numtexts);
	for (int i = 0; i < numtexts; ++i) {
		mpatchbin bin;
		size_t nthsize = getbeuint16((char *)Data + 1 + (2 * i));
		LOG(2) << "text " << i << " is " << nthsize << std::endl;
		char *start = (char *)Data + consumed;
		consumed += nthsize;
		if (consumed > Size) {
			LOG(2) << "ran out of data, consumed " << consumed
			       << " of " << Size << std::endl;
			return 0;
		}
		bin.len = nthsize;
		bin.data.reset(new char[nthsize]);
		memcpy(bin.data.get(), start, nthsize);
		bins.push_back(std::move(bin));
	}
	LOG(2) << "mpatch_flist" << std::endl;
	struct mpatch_flist *patch =
	    mpatch_fold(&bins, getitem, 0, numtexts - 1);
	if (!patch) {
		return 0;
	}
	LOG(2) << "mpatch_calcsize" << std::endl;
	ssize_t outlen = mpatch_calcsize(bins[0].len, patch);
	LOG(2) << "outlen " << outlen << std::endl;
	if (outlen < 0 || outlen > MAX_OUTPUT_SIZE) {
		goto cleanup;
	}
	{
		char *dest = (char *)malloc(outlen);
		LOG(2) << "expecting " << outlen << " total bytes at "
		       << (void *)dest << std::endl;
		mpatch_apply(dest, bins[0].data.get(), bins[0].len, patch);
		free(dest);
		LOG(1) << "applied a complete patch" << std::endl;
	}
cleanup:
	mpatch_lfree(patch);
	return 0;
}

} // extern "C"