revlog: add a `entry_binary` method on index

The revlog index is already responsible for unpacking the binary entry, it would be
simpler to make it responsible for packing them. In practice the C version of
the index is already doing this internally.

We introduce a "entry_binary" method that return the binary version of an
existing revision. The method currently need to also take the revlog header to
deal with the "first revision" special case. We will introduce further refactor
in a later changeset to split that logic out.

Differential Revision: https://phab.mercurial-scm.org/D10508

mercurial/cext/revlog.c

@@ -342,6 +342,38 b' static PyObject *index_get(indexObject *'
 		                     sidedata_offset, sidedata_comp_len);
 	}
 }
+/*
+ * Return the raw binary string representing a revision
+ */
+static PyObject *index_entry_binary(indexObject *self, PyObject *args)
+{
+	long rev;
+	int header;
+	const char *data;
+	char entry[v2_hdrsize];
+
+	Py_ssize_t length = index_length(self);
+
+	if (!PyArg_ParseTuple(args, "lI", &rev, &header)) {
+		return NULL;
+	}
+	if (rev < 0 || rev >= length) {
+		PyErr_Format(PyExc_ValueError, "revlog index out of range: %ld",
+		             rev);
+		return NULL;
+	};
+
+	data = index_deref(self, rev);
+	if (data == NULL)
+		return NULL;
+	if (rev == 0) {
+		// put the header at the start of the first entry
+		memcpy(entry, data, self->hdrsize);
+		putbe32(header, entry);
+		return PyBytes_FromStringAndSize(entry, self->hdrsize);
+	}
+	return PyBytes_FromStringAndSize(data, self->hdrsize);
+}
 
 /*
  * Return the hash of node corresponding to the given rev.
@@ -2859,6 +2891,8 b' static PyMethodDef index_methods[] = {'
     {"shortest", (PyCFunction)index_shortest, METH_VARARGS,
      "find length of shortest hex nodeid of a binary ID"},
     {"stats", (PyCFunction)index_stats, METH_NOARGS, "stats for the index"},
+    {"entry_binary", (PyCFunction)index_entry_binary, METH_VARARGS,
+     "return an entry in binary form"},
     {NULL} /* Sentinel */
 };
 

mercurial/pure/parsers.py

@@ -127,10 +127,24 b' class BaseIndexObject(object):'
             r = (offset_type(0, gettype(r[0])),) + r[1:]
         return r
 
+    def entry_binary(self, rev, header):
+        """return the raw binary string representing a revision"""
+        entry = self[rev]
+        p = revlog_constants.INDEX_ENTRY_V1.pack(*entry)
+        if rev == 0:
+            v_fmt = revlog_constants.INDEX_HEADER
+            v_bin = v_fmt.pack(header)
+            p = v_bin + p[v_fmt.size :]
+        return p
+
 
 class IndexObject(BaseIndexObject):
     def __init__(self, data):
-        assert len(data) % self.entry_size == 0
+        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 = []
@@ -272,6 +286,16 b' class Index2Mixin(object):'
             msg = b"cannot rewrite entries outside of this transaction"
             raise KeyError(msg)
 
+    def entry_binary(self, rev, header):
+        """return the raw binary string representing a revision"""
+        entry = self[rev]
+        p = revlog_constants.INDEX_ENTRY_V2.pack(*entry)
+        if rev == 0:
+            v_fmt = revlog_constants.INDEX_HEADER
+            v_bin = v_fmt.pack(header)
+            p = v_bin + p[v_fmt.size :]
+        return p
+
 
 class IndexObject2(Index2Mixin, IndexObject):
     pass

mercurial/revlog.py

@@ -268,6 +268,24 b' class revlogoldindex(list):'
             return (0, 0, 0, -1, -1, -1, -1, sha1nodeconstants.nullid)
         return list.__getitem__(self, i)
 
+    def entry_binary(self, rev, header):
+        """return the raw binary string representing a revision"""
+        entry = self[rev]
+        if gettype(entry[0]):
+            raise error.RevlogError(
+                _(b'index entry flags need revlog version 1')
+            )
+        e2 = (
+            getoffset(entry[0]),
+            entry[1],
+            entry[3],
+            entry[4],
+            self[entry[5]][7],
+            self[entry[6]][7],
+            entry[7],
+        )
+        return INDEX_ENTRY_V0.pack(*e2)
+
 
 class revlogoldio(object):
     def parseindex(self, data, inline):
@@ -298,29 +316,6 b' class revlogoldio(object):'
         index = revlogoldindex(index)
         return index, None
 
-    def packentry(self, entry, node, version, rev):
-        """return the binary representation of an entry
-
-        entry:   a tuple containing all the values (see index.__getitem__)
-        node:    a callback to convert a revision to nodeid
-        version: the changelog version
-        rev:     the revision number
-        """
-        if gettype(entry[0]):
-            raise error.RevlogError(
-                _(b'index entry flags need revlog version 1')
-            )
-        e2 = (
-            getoffset(entry[0]),
-            entry[1],
-            entry[3],
-            entry[4],
-            node(entry[5]),
-            node(entry[6]),
-            entry[7],
-        )
-        return INDEX_ENTRY_V0.pack(*e2)
-
 
 # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte
 # signed integer)
@@ -333,24 +328,12 b' class revlogio(object):'
         index, cache = parsers.parse_index2(data, inline)
         return index, cache
 
-    def packentry(self, entry, node, version, rev):
-        p = INDEX_ENTRY_V1.pack(*entry)
-        if rev == 0:
-            p = INDEX_HEADER.pack(version) + p[4:]
-        return p
-
 
 class revlogv2io(object):
     def parseindex(self, data, inline):
         index, cache = parsers.parse_index2(data, inline, revlogv2=True)
         return index, cache
 
-    def packentry(self, entry, node, version, rev):
-        p = INDEX_ENTRY_V2.pack(*entry)
-        if rev == 0:
-            p = INDEX_HEADER.pack(version) + p[4:]
-        return p
-
 
 NodemapRevlogIO = None
 
@@ -2068,7 +2051,7 b' class revlog(object):'
             self._inline = False
             io = self._io
             for i in self:
-                e = io.packentry(self.index[i], self.node, self.version, i)
+                e = self.index.entry_binary(i, self.version)
                 fp.write(e)
 
             # the temp file replace the real index when we exit the context
@@ -2390,7 +2373,7 b' class revlog(object):'
             e = e[:8]
 
         self.index.append(e)
-        entry = self._io.packentry(e, self.node, self.version, curr)
+        entry = self.index.entry_binary(curr, self.version)
         self._writeentry(
             transaction,
             ifh,
@@ -3243,5 +3226,5 b' class revlog(object):'
             for i, entry in enumerate(new_entries):
                 rev = startrev + i
                 self.index.replace_sidedata_info(rev, entry[8], entry[9])
-                packed = self._io.packentry(entry, self.node, self.version, rev)
+                packed = self.index.entry_binary(rev, self.version)
                 fp.write(packed)

rust/hg-cpython/src/revlog.rs

@@ -172,6 +172,11 b' py_class!(pub class MixedIndex |py| {'
         self.call_cindex(py, "clearcaches", args, kw)
     }
 
+    /// return the raw binary string representing a revision
+    def entry_binary(&self, *args, **kw) -> PyResult<PyObject> {
+        self.call_cindex(py, "entry_binary", args, kw)
+    }
+
     /// get an index entry
     def get(&self, *args, **kw) -> PyResult<PyObject> {
         self.call_cindex(py, "get", args, kw)