dirstate-entry: introduce dedicated accessors for v1 serialization

In the spirit of changing the content and storage of the dirstate entry, we add
new method that the code doing v1 serialisation can use.

Adding such method to the C object is quite trivial.

Differential Revision: https://phab.mercurial-scm.org/D10951

mercurial/cext/parsers.c

@@ -1,763 +1,795 b''
 /*
  parsers.c - efficient content parsing
 
  Copyright 2008 Olivia Mackall <[e-mail]> and others
 
  This software may be used and distributed according to the terms of
  the GNU General Public License, incorporated herein by reference.
 */
 
 #define PY_SSIZE_T_CLEAN
 #include <Python.h>
 #include <ctype.h>
 #include <stddef.h>
 #include <string.h>
 
 #include "bitmanipulation.h"
 #include "charencode.h"
 #include "util.h"
 
 #ifdef IS_PY3K
 /* The mapping of Python types is meant to be temporary to get Python
  * 3 to compile. We should remove this once Python 3 support is fully
  * supported and proper types are used in the extensions themselves. */
 #define PyInt_Check PyLong_Check
 #define PyInt_FromLong PyLong_FromLong
 #define PyInt_FromSsize_t PyLong_FromSsize_t
 #define PyInt_AsLong PyLong_AsLong
 #endif
 
 static const char *const versionerrortext = "Python minor version mismatch";
 
 static PyObject *dict_new_presized(PyObject *self, PyObject *args)
 {
 	Py_ssize_t expected_size;
 
 	if (!PyArg_ParseTuple(args, "n:make_presized_dict", &expected_size)) {
 		return NULL;
 	}
 
 	return _dict_new_presized(expected_size);
 }
 
 static inline dirstateTupleObject *make_dirstate_tuple(char state, int mode,
                                                        int size, int mtime)
 {
 	dirstateTupleObject *t =
 	    PyObject_New(dirstateTupleObject, &dirstateTupleType);
 	if (!t) {
 		return NULL;
 	}
 	t->state = state;
 	t->mode = mode;
 	t->size = size;
 	t->mtime = mtime;
 	return t;
 }
 
 static PyObject *dirstate_tuple_new(PyTypeObject *subtype, PyObject *args,
                                     PyObject *kwds)
 {
 	/* We do all the initialization here and not a tp_init function because
 	 * dirstate_tuple is immutable. */
 	dirstateTupleObject *t;
 	char state;
 	int size, mode, mtime;
 	if (!PyArg_ParseTuple(args, "ciii", &state, &mode, &size, &mtime)) {
 		return NULL;
 	}
 
 	t = (dirstateTupleObject *)subtype->tp_alloc(subtype, 1);
 	if (!t) {
 		return NULL;
 	}
 	t->state = state;
 	t->mode = mode;
 	t->size = size;
 	t->mtime = mtime;
 
 	return (PyObject *)t;
 }
 
 static void dirstate_tuple_dealloc(PyObject *o)
 {
 	PyObject_Del(o);
 }
 
 static Py_ssize_t dirstate_tuple_length(PyObject *o)
 {
 	return 4;
 }
 
 static PyObject *dirstate_tuple_item(PyObject *o, Py_ssize_t i)
 {
 	dirstateTupleObject *t = (dirstateTupleObject *)o;
 	switch (i) {
 	case 0:
 		return PyBytes_FromStringAndSize(&t->state, 1);
 	case 1:
 		return PyInt_FromLong(t->mode);
 	case 2:
 		return PyInt_FromLong(t->size);
 	case 3:
 		return PyInt_FromLong(t->mtime);
 	default:
 		PyErr_SetString(PyExc_IndexError, "index out of range");
 		return NULL;
 	}
 }
 
 static PySequenceMethods dirstate_tuple_sq = {
     dirstate_tuple_length, /* sq_length */
     0,                     /* sq_concat */
     0,                     /* sq_repeat */
     dirstate_tuple_item,   /* sq_item */
     0,                     /* sq_ass_item */
     0,                     /* sq_contains */
     0,                     /* sq_inplace_concat */
     0                      /* sq_inplace_repeat */
 };
 
+static PyObject *dirstatetuple_v1_state(dirstateTupleObject *self)
+{
+	return PyBytes_FromStringAndSize(&self->state, 1);
+};
+
+static PyObject *dirstatetuple_v1_mode(dirstateTupleObject *self)
+{
+	return PyInt_FromLong(self->mode);
+};
+
+static PyObject *dirstatetuple_v1_size(dirstateTupleObject *self)
+{
+	return PyInt_FromLong(self->size);
+};
+
+static PyObject *dirstatetuple_v1_mtime(dirstateTupleObject *self)
+{
+	return PyInt_FromLong(self->mtime);
+};
+
+static PyMethodDef dirstatetuple_methods[] = {
+    {"v1_state", (PyCFunction)dirstatetuple_v1_state, METH_NOARGS,
+     "return a \"state\" suitable for v1 serialization"},
+    {"v1_mode", (PyCFunction)dirstatetuple_v1_mode, METH_NOARGS,
+     "return a \"mode\" suitable for v1 serialization"},
+    {"v1_size", (PyCFunction)dirstatetuple_v1_size, METH_NOARGS,
+     "return a \"size\" suitable for v1 serialization"},
+    {"v1_mtime", (PyCFunction)dirstatetuple_v1_mtime, METH_NOARGS,
+     "return a \"mtime\" suitable for v1 serialization"},
+    {NULL} /* Sentinel */
+};
+
 PyTypeObject dirstateTupleType = {
     PyVarObject_HEAD_INIT(NULL, 0)      /* header */
     "dirstate_tuple",                   /* tp_name */
     sizeof(dirstateTupleObject),        /* tp_basicsize */
     0,                                  /* tp_itemsize */
     (destructor)dirstate_tuple_dealloc, /* tp_dealloc */
     0,                                  /* tp_print */
     0,                                  /* tp_getattr */
     0,                                  /* tp_setattr */
     0,                                  /* tp_compare */
     0,                                  /* tp_repr */
     0,                                  /* tp_as_number */
     &dirstate_tuple_sq,                 /* tp_as_sequence */
     0,                                  /* tp_as_mapping */
     0,                                  /* tp_hash  */
     0,                                  /* tp_call */
     0,                                  /* tp_str */
     0,                                  /* tp_getattro */
     0,                                  /* tp_setattro */
     0,                                  /* tp_as_buffer */
     Py_TPFLAGS_DEFAULT,                 /* tp_flags */
     "dirstate tuple",                   /* tp_doc */
     0,                                  /* tp_traverse */
     0,                                  /* tp_clear */
     0,                                  /* tp_richcompare */
     0,                                  /* tp_weaklistoffset */
     0,                                  /* tp_iter */
     0,                                  /* tp_iternext */
-    0,                                  /* tp_methods */
+    dirstatetuple_methods,              /* tp_methods */
     0,                                  /* tp_members */
     0,                                  /* tp_getset */
     0,                                  /* tp_base */
     0,                                  /* tp_dict */
     0,                                  /* tp_descr_get */
     0,                                  /* tp_descr_set */
     0,                                  /* tp_dictoffset */
     0,                                  /* tp_init */
     0,                                  /* tp_alloc */
     dirstate_tuple_new,                 /* tp_new */
 };
 
 static PyObject *parse_dirstate(PyObject *self, PyObject *args)
 {
 	PyObject *dmap, *cmap, *parents = NULL, *ret = NULL;
 	PyObject *fname = NULL, *cname = NULL, *entry = NULL;
 	char state, *cur, *str, *cpos;
 	int mode, size, mtime;
 	unsigned int flen, pos = 40;
 	Py_ssize_t len = 40;
 	Py_ssize_t readlen;
 
 	if (!PyArg_ParseTuple(
 	        args, PY23("O!O!s#:parse_dirstate", "O!O!y#:parse_dirstate"),
 	        &PyDict_Type, &dmap, &PyDict_Type, &cmap, &str, &readlen)) {
 		goto quit;
 	}
 
 	len = readlen;
 
 	/* read parents */
 	if (len < 40) {
 		PyErr_SetString(PyExc_ValueError,
 		                "too little data for parents");
 		goto quit;
 	}
 
 	parents = Py_BuildValue(PY23("s#s#", "y#y#"), str, (Py_ssize_t)20,
 	                        str + 20, (Py_ssize_t)20);
 	if (!parents) {
 		goto quit;
 	}
 
 	/* read filenames */
 	while (pos >= 40 && pos < len) {
 		if (pos + 17 > len) {
 			PyErr_SetString(PyExc_ValueError,
 			                "overflow in dirstate");
 			goto quit;
 		}
 		cur = str + pos;
 		/* unpack header */
 		state = *cur;
 		mode = getbe32(cur + 1);
 		size = getbe32(cur + 5);
 		mtime = getbe32(cur + 9);
 		flen = getbe32(cur + 13);
 		pos += 17;
 		cur += 17;
 		if (flen > len - pos) {
 			PyErr_SetString(PyExc_ValueError,
 			                "overflow in dirstate");
 			goto quit;
 		}
 
 		entry =
 		    (PyObject *)make_dirstate_tuple(state, mode, size, mtime);
 		cpos = memchr(cur, 0, flen);
 		if (cpos) {
 			fname = PyBytes_FromStringAndSize(cur, cpos - cur);
 			cname = PyBytes_FromStringAndSize(
 			    cpos + 1, flen - (cpos - cur) - 1);
 			if (!fname || !cname ||
 			    PyDict_SetItem(cmap, fname, cname) == -1 ||
 			    PyDict_SetItem(dmap, fname, entry) == -1) {
 				goto quit;
 			}
 			Py_DECREF(cname);
 		} else {
 			fname = PyBytes_FromStringAndSize(cur, flen);
 			if (!fname ||
 			    PyDict_SetItem(dmap, fname, entry) == -1) {
 				goto quit;
 			}
 		}
 		Py_DECREF(fname);
 		Py_DECREF(entry);
 		fname = cname = entry = NULL;
 		pos += flen;
 	}
 
 	ret = parents;
 	Py_INCREF(ret);
 quit:
 	Py_XDECREF(fname);
 	Py_XDECREF(cname);
 	Py_XDECREF(entry);
 	Py_XDECREF(parents);
 	return ret;
 }
 
 /*
  * Build a set of non-normal and other parent entries from the dirstate dmap
  */
 static PyObject *nonnormalotherparententries(PyObject *self, PyObject *args)
 {
 	PyObject *dmap, *fname, *v;
 	PyObject *nonnset = NULL, *otherpset = NULL, *result = NULL;
 	Py_ssize_t pos;
 
 	if (!PyArg_ParseTuple(args, "O!:nonnormalentries", &PyDict_Type,
 	                      &dmap)) {
 		goto bail;
 	}
 
 	nonnset = PySet_New(NULL);
 	if (nonnset == NULL) {
 		goto bail;
 	}
 
 	otherpset = PySet_New(NULL);
 	if (otherpset == NULL) {
 		goto bail;
 	}
 
 	pos = 0;
 	while (PyDict_Next(dmap, &pos, &fname, &v)) {
 		dirstateTupleObject *t;
 		if (!dirstate_tuple_check(v)) {
 			PyErr_SetString(PyExc_TypeError,
 			                "expected a dirstate tuple");
 			goto bail;
 		}
 		t = (dirstateTupleObject *)v;
 
 		if (t->state == 'n' && t->size == -2) {
 			if (PySet_Add(otherpset, fname) == -1) {
 				goto bail;
 			}
 		}
 
 		if (t->state == 'n' && t->mtime != -1) {
 			continue;
 		}
 		if (PySet_Add(nonnset, fname) == -1) {
 			goto bail;
 		}
 	}
 
 	result = Py_BuildValue("(OO)", nonnset, otherpset);
 	if (result == NULL) {
 		goto bail;
 	}
 	Py_DECREF(nonnset);
 	Py_DECREF(otherpset);
 	return result;
 bail:
 	Py_XDECREF(nonnset);
 	Py_XDECREF(otherpset);
 	Py_XDECREF(result);
 	return NULL;
 }
 
 /*
  * Efficiently pack a dirstate object into its on-disk format.
  */
 static PyObject *pack_dirstate(PyObject *self, PyObject *args)
 {
 	PyObject *packobj = NULL;
 	PyObject *map, *copymap, *pl, *mtime_unset = NULL;
 	Py_ssize_t nbytes, pos, l;
 	PyObject *k, *v = NULL, *pn;
 	char *p, *s;
 	int now;
 
 	if (!PyArg_ParseTuple(args, "O!O!O!i:pack_dirstate", &PyDict_Type, &map,
 	                      &PyDict_Type, &copymap, &PyTuple_Type, &pl,
 	                      &now)) {
 		return NULL;
 	}
 
 	if (PyTuple_Size(pl) != 2) {
 		PyErr_SetString(PyExc_TypeError, "expected 2-element tuple");
 		return NULL;
 	}
 
 	/* Figure out how much we need to allocate. */
 	for (nbytes = 40, pos = 0; PyDict_Next(map, &pos, &k, &v);) {
 		PyObject *c;
 		if (!PyBytes_Check(k)) {
 			PyErr_SetString(PyExc_TypeError, "expected string key");
 			goto bail;
 		}
 		nbytes += PyBytes_GET_SIZE(k) + 17;
 		c = PyDict_GetItem(copymap, k);
 		if (c) {
 			if (!PyBytes_Check(c)) {
 				PyErr_SetString(PyExc_TypeError,
 				                "expected string key");
 				goto bail;
 			}
 			nbytes += PyBytes_GET_SIZE(c) + 1;
 		}
 	}
 
 	packobj = PyBytes_FromStringAndSize(NULL, nbytes);
 	if (packobj == NULL) {
 		goto bail;
 	}
 
 	p = PyBytes_AS_STRING(packobj);
 
 	pn = PyTuple_GET_ITEM(pl, 0);
 	if (PyBytes_AsStringAndSize(pn, &s, &l) == -1 || l != 20) {
 		PyErr_SetString(PyExc_TypeError, "expected a 20-byte hash");
 		goto bail;
 	}
 	memcpy(p, s, l);
 	p += 20;
 	pn = PyTuple_GET_ITEM(pl, 1);
 	if (PyBytes_AsStringAndSize(pn, &s, &l) == -1 || l != 20) {
 		PyErr_SetString(PyExc_TypeError, "expected a 20-byte hash");
 		goto bail;
 	}
 	memcpy(p, s, l);
 	p += 20;
 
 	for (pos = 0; PyDict_Next(map, &pos, &k, &v);) {
 		dirstateTupleObject *tuple;
 		char state;
 		int mode, size, mtime;
 		Py_ssize_t len, l;
 		PyObject *o;
 		char *t;
 
 		if (!dirstate_tuple_check(v)) {
 			PyErr_SetString(PyExc_TypeError,
 			                "expected a dirstate tuple");
 			goto bail;
 		}
 		tuple = (dirstateTupleObject *)v;
 
 		state = tuple->state;
 		mode = tuple->mode;
 		size = tuple->size;
 		mtime = tuple->mtime;
 		if (state == 'n' && mtime == now) {
 			/* See pure/parsers.py:pack_dirstate for why we do
 			 * this. */
 			mtime = -1;
 			mtime_unset = (PyObject *)make_dirstate_tuple(
 			    state, mode, size, mtime);
 			if (!mtime_unset) {
 				goto bail;
 			}
 			if (PyDict_SetItem(map, k, mtime_unset) == -1) {
 				goto bail;
 			}
 			Py_DECREF(mtime_unset);
 			mtime_unset = NULL;
 		}
 		*p++ = state;
 		putbe32((uint32_t)mode, p);
 		putbe32((uint32_t)size, p + 4);
 		putbe32((uint32_t)mtime, p + 8);
 		t = p + 12;
 		p += 16;
 		len = PyBytes_GET_SIZE(k);
 		memcpy(p, PyBytes_AS_STRING(k), len);
 		p += len;
 		o = PyDict_GetItem(copymap, k);
 		if (o) {
 			*p++ = '\0';
 			l = PyBytes_GET_SIZE(o);
 			memcpy(p, PyBytes_AS_STRING(o), l);
 			p += l;
 			len += l + 1;
 		}
 		putbe32((uint32_t)len, t);
 	}
 
 	pos = p - PyBytes_AS_STRING(packobj);
 	if (pos != nbytes) {
 		PyErr_Format(PyExc_SystemError, "bad dirstate size: %ld != %ld",
 		             (long)pos, (long)nbytes);
 		goto bail;
 	}
 
 	return packobj;
 bail:
 	Py_XDECREF(mtime_unset);
 	Py_XDECREF(packobj);
 	Py_XDECREF(v);
 	return NULL;
 }
 
 #define BUMPED_FIX 1
 #define USING_SHA_256 2
 #define FM1_HEADER_SIZE (4 + 8 + 2 + 2 + 1 + 1 + 1)
 
 static PyObject *readshas(const char *source, unsigned char num,
                           Py_ssize_t hashwidth)
 {
 	int i;
 	PyObject *list = PyTuple_New(num);
 	if (list == NULL) {
 		return NULL;
 	}
 	for (i = 0; i < num; i++) {
 		PyObject *hash = PyBytes_FromStringAndSize(source, hashwidth);
 		if (hash == NULL) {
 			Py_DECREF(list);
 			return NULL;
 		}
 		PyTuple_SET_ITEM(list, i, hash);
 		source += hashwidth;
 	}
 	return list;
 }
 
 static PyObject *fm1readmarker(const char *databegin, const char *dataend,
                                uint32_t *msize)
 {
 	const char *data = databegin;
 	const char *meta;
 
 	double mtime;
 	int16_t tz;
 	uint16_t flags;
 	unsigned char nsuccs, nparents, nmetadata;
 	Py_ssize_t hashwidth = 20;
 
 	PyObject *prec = NULL, *parents = NULL, *succs = NULL;
 	PyObject *metadata = NULL, *ret = NULL;
 	int i;
 
 	if (data + FM1_HEADER_SIZE > dataend) {
 		goto overflow;
 	}
 
 	*msize = getbe32(data);
 	data += 4;
 	mtime = getbefloat64(data);
 	data += 8;
 	tz = getbeint16(data);
 	data += 2;
 	flags = getbeuint16(data);
 	data += 2;
 
 	if (flags & USING_SHA_256) {
 		hashwidth = 32;
 	}
 
 	nsuccs = (unsigned char)(*data++);
 	nparents = (unsigned char)(*data++);
 	nmetadata = (unsigned char)(*data++);
 
 	if (databegin + *msize > dataend) {
 		goto overflow;
 	}
 	dataend = databegin + *msize; /* narrow down to marker size */
 
 	if (data + hashwidth > dataend) {
 		goto overflow;
 	}
 	prec = PyBytes_FromStringAndSize(data, hashwidth);
 	data += hashwidth;
 	if (prec == NULL) {
 		goto bail;
 	}
 
 	if (data + nsuccs * hashwidth > dataend) {
 		goto overflow;
 	}
 	succs = readshas(data, nsuccs, hashwidth);
 	if (succs == NULL) {
 		goto bail;
 	}
 	data += nsuccs * hashwidth;
 
 	if (nparents == 1 || nparents == 2) {
 		if (data + nparents * hashwidth > dataend) {
 			goto overflow;
 		}
 		parents = readshas(data, nparents, hashwidth);
 		if (parents == NULL) {
 			goto bail;
 		}
 		data += nparents * hashwidth;
 	} else {
 		parents = Py_None;
 		Py_INCREF(parents);
 	}
 
 	if (data + 2 * nmetadata > dataend) {
 		goto overflow;
 	}
 	meta = data + (2 * nmetadata);
 	metadata = PyTuple_New(nmetadata);
 	if (metadata == NULL) {
 		goto bail;
 	}
 	for (i = 0; i < nmetadata; i++) {
 		PyObject *tmp, *left = NULL, *right = NULL;
 		Py_ssize_t leftsize = (unsigned char)(*data++);
 		Py_ssize_t rightsize = (unsigned char)(*data++);
 		if (meta + leftsize + rightsize > dataend) {
 			goto overflow;
 		}
 		left = PyBytes_FromStringAndSize(meta, leftsize);
 		meta += leftsize;
 		right = PyBytes_FromStringAndSize(meta, rightsize);
 		meta += rightsize;
 		tmp = PyTuple_New(2);
 		if (!left || !right || !tmp) {
 			Py_XDECREF(left);
 			Py_XDECREF(right);
 			Py_XDECREF(tmp);
 			goto bail;
 		}
 		PyTuple_SET_ITEM(tmp, 0, left);
 		PyTuple_SET_ITEM(tmp, 1, right);
 		PyTuple_SET_ITEM(metadata, i, tmp);
 	}
 	ret = Py_BuildValue("(OOHO(di)O)", prec, succs, flags, metadata, mtime,
 	                    (int)tz * 60, parents);
 	goto bail; /* return successfully */
 
 overflow:
 	PyErr_SetString(PyExc_ValueError, "overflow in obsstore");
 bail:
 	Py_XDECREF(prec);
 	Py_XDECREF(succs);
 	Py_XDECREF(metadata);
 	Py_XDECREF(parents);
 	return ret;
 }
 
 static PyObject *fm1readmarkers(PyObject *self, PyObject *args)
 {
 	const char *data, *dataend;
 	Py_ssize_t datalen, offset, stop;
 	PyObject *markers = NULL;
 
 	if (!PyArg_ParseTuple(args, PY23("s#nn", "y#nn"), &data, &datalen,
 	                      &offset, &stop)) {
 		return NULL;
 	}
 	if (offset < 0) {
 		PyErr_SetString(PyExc_ValueError,
 		                "invalid negative offset in fm1readmarkers");
 		return NULL;
 	}
 	if (stop > datalen) {
 		PyErr_SetString(
 		    PyExc_ValueError,
 		    "stop longer than data length in fm1readmarkers");
 		return NULL;
 	}
 	dataend = data + datalen;
 	data += offset;
 	markers = PyList_New(0);
 	if (!markers) {
 		return NULL;
 	}
 	while (offset < stop) {
 		uint32_t msize;
 		int error;
 		PyObject *record = fm1readmarker(data, dataend, &msize);
 		if (!record) {
 			goto bail;
 		}
 		error = PyList_Append(markers, record);
 		Py_DECREF(record);
 		if (error) {
 			goto bail;
 		}
 		data += msize;
 		offset += msize;
 	}
 	return markers;
 bail:
 	Py_DECREF(markers);
 	return NULL;
 }
 
 static char parsers_doc[] = "Efficient content parsing.";
 
 PyObject *encodedir(PyObject *self, PyObject *args);
 PyObject *pathencode(PyObject *self, PyObject *args);
 PyObject *lowerencode(PyObject *self, PyObject *args);
 PyObject *parse_index2(PyObject *self, PyObject *args, PyObject *kwargs);
 
 static PyMethodDef methods[] = {
     {"pack_dirstate", pack_dirstate, METH_VARARGS, "pack a dirstate\n"},
     {"nonnormalotherparententries", nonnormalotherparententries, METH_VARARGS,
      "create a set containing non-normal and other parent entries of given "
      "dirstate\n"},
     {"parse_dirstate", parse_dirstate, METH_VARARGS, "parse a dirstate\n"},
     {"parse_index2", (PyCFunction)parse_index2, METH_VARARGS | METH_KEYWORDS,
      "parse a revlog index\n"},
     {"isasciistr", isasciistr, METH_VARARGS, "check if an ASCII string\n"},
     {"asciilower", asciilower, METH_VARARGS, "lowercase an ASCII string\n"},
     {"asciiupper", asciiupper, METH_VARARGS, "uppercase an ASCII string\n"},
     {"dict_new_presized", dict_new_presized, METH_VARARGS,
      "construct a dict with an expected size\n"},
     {"make_file_foldmap", make_file_foldmap, METH_VARARGS,
      "make file foldmap\n"},
     {"jsonescapeu8fast", jsonescapeu8fast, METH_VARARGS,
      "escape a UTF-8 byte string to JSON (fast path)\n"},
     {"encodedir", encodedir, METH_VARARGS, "encodedir a path\n"},
     {"pathencode", pathencode, METH_VARARGS, "fncache-encode a path\n"},
     {"lowerencode", lowerencode, METH_VARARGS, "lower-encode a path\n"},
     {"fm1readmarkers", fm1readmarkers, METH_VARARGS,
      "parse v1 obsolete markers\n"},
     {NULL, NULL}};
 
 void dirs_module_init(PyObject *mod);
 void manifest_module_init(PyObject *mod);
 void revlog_module_init(PyObject *mod);
 
 static const int version = 20;
 
 static void module_init(PyObject *mod)
 {
 	PyObject *capsule = NULL;
 	PyModule_AddIntConstant(mod, "version", version);
 
 	/* This module constant has two purposes.  First, it lets us unit test
 	 * the ImportError raised without hard-coding any error text.  This
 	 * means we can change the text in the future without breaking tests,
 	 * even across changesets without a recompile.  Second, its presence
 	 * can be used to determine whether the version-checking logic is
 	 * present, which also helps in testing across changesets without a
 	 * recompile.  Note that this means the pure-Python version of parsers
 	 * should not have this module constant. */
 	PyModule_AddStringConstant(mod, "versionerrortext", versionerrortext);
 
 	dirs_module_init(mod);
 	manifest_module_init(mod);
 	revlog_module_init(mod);
 
 	capsule = PyCapsule_New(
 	    make_dirstate_tuple,
 	    "mercurial.cext.parsers.make_dirstate_tuple_CAPI", NULL);
 	if (capsule != NULL)
 		PyModule_AddObject(mod, "make_dirstate_tuple_CAPI", capsule);
 
 	if (PyType_Ready(&dirstateTupleType) < 0) {
 		return;
 	}
 	Py_INCREF(&dirstateTupleType);
 	PyModule_AddObject(mod, "dirstatetuple",
 	                   (PyObject *)&dirstateTupleType);
 }
 
 static int check_python_version(void)
 {
 	PyObject *sys = PyImport_ImportModule("sys"), *ver;
 	long hexversion;
 	if (!sys) {
 		return -1;
 	}
 	ver = PyObject_GetAttrString(sys, "hexversion");
 	Py_DECREF(sys);
 	if (!ver) {
 		return -1;
 	}
 	hexversion = PyInt_AsLong(ver);
 	Py_DECREF(ver);
 	/* sys.hexversion is a 32-bit number by default, so the -1 case
 	 * should only occur in unusual circumstances (e.g. if sys.hexversion
 	 * is manually set to an invalid value). */
 	if ((hexversion == -1) || (hexversion >> 16 != PY_VERSION_HEX >> 16)) {
 		PyErr_Format(PyExc_ImportError,
 		             "%s: The Mercurial extension "
 		             "modules were compiled with Python " PY_VERSION
 		             ", but "
 		             "Mercurial is currently using Python with "
 		             "sys.hexversion=%ld: "
 		             "Python %s\n at: %s",
 		             versionerrortext, hexversion, Py_GetVersion(),
 		             Py_GetProgramFullPath());
 		return -1;
 	}
 	return 0;
 }
 
 #ifdef IS_PY3K
 static struct PyModuleDef parsers_module = {PyModuleDef_HEAD_INIT, "parsers",
                                             parsers_doc, -1, methods};
 
 PyMODINIT_FUNC PyInit_parsers(void)
 {
 	PyObject *mod;
 
 	if (check_python_version() == -1)
 		return NULL;
 	mod = PyModule_Create(&parsers_module);
 	module_init(mod);
 	return mod;
 }
 #else
 PyMODINIT_FUNC initparsers(void)
 {
 	PyObject *mod;
 
 	if (check_python_version() == -1) {
 		return;
 	}
 	mod = Py_InitModule3("parsers", methods, parsers_doc);
 	module_init(mod);
 }
 #endif

mercurial/pure/parsers.py

@@ -1,456 +1,479 b''
 # parsers.py - Python implementation of parsers.c
 #
 # Copyright 2009 Olivia Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import struct
 import zlib
 
 from ..node import (
     nullrev,
     sha1nodeconstants,
 )
 from .. import (
     error,
     pycompat,
     revlogutils,
     util,
 )
 
 from ..revlogutils import nodemap as nodemaputil
 from ..revlogutils import constants as revlog_constants
 
 stringio = pycompat.bytesio
 
 
 _pack = struct.pack
 _unpack = struct.unpack
 _compress = zlib.compress
 _decompress = zlib.decompress
 
 
 class dirstatetuple(object):
     """represent a dirstate entry
 
     It contains:
 
     - state (one of 'n', 'a', 'r', 'm')
     - mode,
     - size,
     - mtime,
     """
 
     __slot__ = ('_state', '_mode', '_size', '_mtime')
 
     def __init__(self, state, mode, size, mtime):
         self._state = state
         self._mode = mode
         self._size = size
         self._mtime = mtime
 
     def __getitem__(self, idx):
         if idx == 0 or idx == -4:
             return self._state
         elif idx == 1 or idx == -3:
             return self._mode
         elif idx == 2 or idx == -2:
             return self._size
         elif idx == 3 or idx == -1:
             return self._mtime
         else:
             raise IndexError(idx)
 
+    def v1_state(self):
+        """return a "state" suitable for v1 serialization"""
+        return self._state
+
+    def v1_mode(self):
+        """return a "mode" suitable for v1 serialization"""
+        return self._mode
+
+    def v1_size(self):
+        """return a "size" suitable for v1 serialization"""
+        return self._size
+
+    def v1_mtime(self):
+        """return a "mtime" suitable for v1 serialization"""
+        return self._mtime
+
 
 def gettype(q):
     return int(q & 0xFFFF)
 
 
 class BaseIndexObject(object):
     # Can I be passed to an algorithme implemented in Rust ?
     rust_ext_compat = 0
     # Format of an index entry according to Python's `struct` language
     index_format = revlog_constants.INDEX_ENTRY_V1
     # Size of a C unsigned long long int, platform independent
     big_int_size = struct.calcsize(b'>Q')
     # Size of a C long int, platform independent
     int_size = struct.calcsize(b'>i')
     # An empty index entry, used as a default value to be overridden, or nullrev
     null_item = (
         0,
         0,
         0,
         -1,
         -1,
         -1,
         -1,
         sha1nodeconstants.nullid,
         0,
         0,
         revlog_constants.COMP_MODE_INLINE,
         revlog_constants.COMP_MODE_INLINE,
     )
 
     @util.propertycache
     def entry_size(self):
         return self.index_format.size
 
     @property
     def nodemap(self):
         msg = b"index.nodemap is deprecated, use index.[has_node|rev|get_rev]"
         util.nouideprecwarn(msg, b'5.3', stacklevel=2)
         return self._nodemap
 
     @util.propertycache
     def _nodemap(self):
         nodemap = nodemaputil.NodeMap({sha1nodeconstants.nullid: nullrev})
         for r in range(0, len(self)):
             n = self[r][7]
             nodemap[n] = r
         return nodemap
 
     def has_node(self, node):
         """return True if the node exist in the index"""
         return node in self._nodemap
 
     def rev(self, node):
         """return a revision for a node
 
         If the node is unknown, raise a RevlogError"""
         return self._nodemap[node]
 
     def get_rev(self, node):
         """return a revision for a node
 
         If the node is unknown, return None"""
         return self._nodemap.get(node)
 
     def _stripnodes(self, start):
         if '_nodemap' in vars(self):
             for r in range(start, len(self)):
                 n = self[r][7]
                 del self._nodemap[n]
 
     def clearcaches(self):
         self.__dict__.pop('_nodemap', None)
 
     def __len__(self):
         return self._lgt + len(self._extra)
 
     def append(self, tup):
         if '_nodemap' in vars(self):
             self._nodemap[tup[7]] = len(self)
         data = self._pack_entry(len(self), tup)
         self._extra.append(data)
 
     def _pack_entry(self, rev, entry):
         assert entry[8] == 0
         assert entry[9] == 0
         return self.index_format.pack(*entry[:8])
 
     def _check_index(self, i):
         if not isinstance(i, int):
             raise TypeError(b"expecting int indexes")
         if i < 0 or i >= len(self):
             raise IndexError
 
     def __getitem__(self, i):
         if i == -1:
             return self.null_item
         self._check_index(i)
         if i >= self._lgt:
             data = self._extra[i - self._lgt]
         else:
             index = self._calculate_index(i)
             data = self._data[index : index + self.entry_size]
         r = self._unpack_entry(i, data)
         if self._lgt and i == 0:
             offset = revlogutils.offset_type(0, gettype(r[0]))
             r = (offset,) + r[1:]
         return r
 
     def _unpack_entry(self, rev, data):
         r = self.index_format.unpack(data)
         r = r + (
             0,
             0,
             revlog_constants.COMP_MODE_INLINE,
             revlog_constants.COMP_MODE_INLINE,
         )
         return r
 
     def pack_header(self, header):
         """pack header information as binary"""
         v_fmt = revlog_constants.INDEX_HEADER
         return v_fmt.pack(header)
 
     def entry_binary(self, rev):
         """return the raw binary string representing a revision"""
         entry = self[rev]
         p = revlog_constants.INDEX_ENTRY_V1.pack(*entry[:8])
         if rev == 0:
             p = p[revlog_constants.INDEX_HEADER.size :]
         return p
 
 
 class IndexObject(BaseIndexObject):
     def __init__(self, data):
         assert len(data) % self.entry_size == 0, (
             len(data),
             self.entry_size,
             len(data) % self.entry_size,
         )
         self._data = data
         self._lgt = len(data) // self.entry_size
         self._extra = []
 
     def _calculate_index(self, i):
         return i * self.entry_size
 
     def __delitem__(self, i):
         if not isinstance(i, slice) or not i.stop == -1 or i.step is not None:
             raise ValueError(b"deleting slices only supports a:-1 with step 1")
         i = i.start
         self._check_index(i)
         self._stripnodes(i)
         if i < self._lgt:
             self._data = self._data[: i * self.entry_size]
             self._lgt = i
             self._extra = []
         else:
             self._extra = self._extra[: i - self._lgt]
 
 
 class PersistentNodeMapIndexObject(IndexObject):
     """a Debug oriented class to test persistent nodemap
 
     We need a simple python object to test API and higher level behavior. See
     the Rust implementation for  more serious usage. This should be used only
     through the dedicated `devel.persistent-nodemap` config.
     """
 
     def nodemap_data_all(self):
         """Return bytes containing a full serialization of a nodemap
 
         The nodemap should be valid for the full set of revisions in the
         index."""
         return nodemaputil.persistent_data(self)
 
     def nodemap_data_incremental(self):
         """Return bytes containing a incremental update to persistent nodemap
 
         This containst the data for an append-only update of the data provided
         in the last call to `update_nodemap_data`.
         """
         if self._nm_root is None:
             return None
         docket = self._nm_docket
         changed, data = nodemaputil.update_persistent_data(
             self, self._nm_root, self._nm_max_idx, self._nm_docket.tip_rev
         )
 
         self._nm_root = self._nm_max_idx = self._nm_docket = None
         return docket, changed, data
 
     def update_nodemap_data(self, docket, nm_data):
         """provide full block of persisted binary data for a nodemap
 
         The data are expected to come from disk. See `nodemap_data_all` for a
         produceur of such data."""
         if nm_data is not None:
             self._nm_root, self._nm_max_idx = nodemaputil.parse_data(nm_data)
             if self._nm_root:
                 self._nm_docket = docket
             else:
                 self._nm_root = self._nm_max_idx = self._nm_docket = None
 
 
 class InlinedIndexObject(BaseIndexObject):
     def __init__(self, data, inline=0):
         self._data = data
         self._lgt = self._inline_scan(None)
         self._inline_scan(self._lgt)
         self._extra = []
 
     def _inline_scan(self, lgt):
         off = 0
         if lgt is not None:
             self._offsets = [0] * lgt
         count = 0
         while off <= len(self._data) - self.entry_size:
             start = off + self.big_int_size
             (s,) = struct.unpack(
                 b'>i',
                 self._data[start : start + self.int_size],
             )
             if lgt is not None:
                 self._offsets[count] = off
             count += 1
             off += self.entry_size + s
         if off != len(self._data):
             raise ValueError(b"corrupted data")
         return count
 
     def __delitem__(self, i):
         if not isinstance(i, slice) or not i.stop == -1 or i.step is not None:
             raise ValueError(b"deleting slices only supports a:-1 with step 1")
         i = i.start
         self._check_index(i)
         self._stripnodes(i)
         if i < self._lgt:
             self._offsets = self._offsets[:i]
             self._lgt = i
             self._extra = []
         else:
             self._extra = self._extra[: i - self._lgt]
 
     def _calculate_index(self, i):
         return self._offsets[i]
 
 
 def parse_index2(data, inline, revlogv2=False):
     if not inline:
         cls = IndexObject2 if revlogv2 else IndexObject
         return cls(data), None
     cls = InlinedIndexObject
     return cls(data, inline), (0, data)
 
 
 def parse_index_cl_v2(data):
     return IndexChangelogV2(data), None
 
 
 class IndexObject2(IndexObject):
     index_format = revlog_constants.INDEX_ENTRY_V2
 
     def replace_sidedata_info(
         self,
         rev,
         sidedata_offset,
         sidedata_length,
         offset_flags,
         compression_mode,
     ):
         """
         Replace an existing index entry's sidedata offset and length with new
         ones.
         This cannot be used outside of the context of sidedata rewriting,
         inside the transaction that creates the revision `rev`.
         """
         if rev < 0:
             raise KeyError
         self._check_index(rev)
         if rev < self._lgt:
             msg = b"cannot rewrite entries outside of this transaction"
             raise KeyError(msg)
         else:
             entry = list(self[rev])
             entry[0] = offset_flags
             entry[8] = sidedata_offset
             entry[9] = sidedata_length
             entry[11] = compression_mode
             entry = tuple(entry)
             new = self._pack_entry(rev, entry)
             self._extra[rev - self._lgt] = new
 
     def _unpack_entry(self, rev, data):
         data = self.index_format.unpack(data)
         entry = data[:10]
         data_comp = data[10] & 3
         sidedata_comp = (data[10] & (3 << 2)) >> 2
         return entry + (data_comp, sidedata_comp)
 
     def _pack_entry(self, rev, entry):
         data = entry[:10]
         data_comp = entry[10] & 3
         sidedata_comp = (entry[11] & 3) << 2
         data += (data_comp | sidedata_comp,)
 
         return self.index_format.pack(*data)
 
     def entry_binary(self, rev):
         """return the raw binary string representing a revision"""
         entry = self[rev]
         return self._pack_entry(rev, entry)
 
     def pack_header(self, header):
         """pack header information as binary"""
         msg = 'version header should go in the docket, not the index: %d'
         msg %= header
         raise error.ProgrammingError(msg)
 
 
 class IndexChangelogV2(IndexObject2):
     index_format = revlog_constants.INDEX_ENTRY_CL_V2
 
     def _unpack_entry(self, rev, data, r=True):
         items = self.index_format.unpack(data)
         entry = items[:3] + (rev, rev) + items[3:8]
         data_comp = items[8] & 3
         sidedata_comp = (items[8] >> 2) & 3
         return entry + (data_comp, sidedata_comp)
 
     def _pack_entry(self, rev, entry):
         assert entry[3] == rev, entry[3]
         assert entry[4] == rev, entry[4]
         data = entry[:3] + entry[5:10]
         data_comp = entry[10] & 3
         sidedata_comp = (entry[11] & 3) << 2
         data += (data_comp | sidedata_comp,)
         return self.index_format.pack(*data)
 
 
 def parse_index_devel_nodemap(data, inline):
     """like parse_index2, but alway return a PersistentNodeMapIndexObject"""
     return PersistentNodeMapIndexObject(data), None
 
 
 def parse_dirstate(dmap, copymap, st):
     parents = [st[:20], st[20:40]]
     # dereference fields so they will be local in loop
     format = b">cllll"
     e_size = struct.calcsize(format)
     pos1 = 40
     l = len(st)
 
     # the inner loop
     while pos1 < l:
         pos2 = pos1 + e_size
         e = _unpack(b">cllll", st[pos1:pos2])  # a literal here is faster
         pos1 = pos2 + e[4]
         f = st[pos2:pos1]
         if b'\0' in f:
             f, c = f.split(b'\0')
             copymap[f] = c
         dmap[f] = dirstatetuple(*e[:4])
     return parents
 
 
 def pack_dirstate(dmap, copymap, pl, now):
     now = int(now)
     cs = stringio()
     write = cs.write
     write(b"".join(pl))
     for f, e in pycompat.iteritems(dmap):
         if e[0] == b'n' and e[3] == now:
             # The file was last modified "simultaneously" with the current
             # write to dirstate (i.e. within the same second for file-
             # systems with a granularity of 1 sec). This commonly happens
             # for at least a couple of files on 'update'.
             # The user could change the file without changing its size
             # within the same second. Invalidate the file's mtime in
             # dirstate, forcing future 'status' calls to compare the
             # contents of the file if the size is the same. This prevents
             # mistakenly treating such files as clean.
             e = dirstatetuple(e[0], e[1], e[2], -1)
             dmap[f] = e
 
         if f in copymap:
             f = b"%s\0%s" % (f, copymap[f])
-        e = _pack(b">cllll", e[0], e[1], e[2], e[3], len(f))
+        e = _pack(
+            b">cllll",
+            e.v1_state(),
+            e.v1_mode(),
+            e.v1_size(),
+            e.v1_mtime(),
+            len(f),
+        )
         write(e)
         write(f)
     return cs.getvalue()