hgext/remotefilelog/metadatastore.py
author Ian Moody <moz-ian@perix.co.uk>
Tue, 16 Jul 2019 19:18:16 +0100
changeset 42618 c17e6a3e7356
parent 40610 13d4ad8d7801
child 43076 2372284d9457
permissions -rw-r--r--
phabricator: handle local:commits time being string or int When setting local:commits arcanist has different behaviour depending on whether the repo is git or hg. With hg it sets the time as a number, since it calls PHP's strtotime on the value, but with git it sets it as a string. Normally this wouldn't be an issue since phabread wouldn't be interacting with Phabricator Revisions for git repos, but Mozilla has a secondary workflow for git users that uses the git-cinnabar tool to interact with their hg repos. When a git-cinnabar user uses the moz-phab tool to submit patches for mozilla-central it makes use of Mozilla's fork of arcanist, which works with their local git version of m-c, and thus sets the local:commit time as a string, and then translates the commit hashes. Currently when encountering such DREVS phabread dies with "TypeError: %d format: a number is required, not str". phabsend also used to set it as a string but wouldn't have encountered the issue with its own DREVs since it would read hg:meta first. Differential Revision: https://phab.mercurial-scm.org/D6650

from __future__ import absolute_import

from mercurial.node import hex, nullid
from . import (
    basestore,
    shallowutil,
)

class unionmetadatastore(basestore.baseunionstore):
    def __init__(self, *args, **kwargs):
        super(unionmetadatastore, self).__init__(*args, **kwargs)

        self.stores = args
        self.writestore = kwargs.get(r'writestore')

        # If allowincomplete==True then the union store can return partial
        # ancestor lists, otherwise it will throw a KeyError if a full
        # history can't be found.
        self.allowincomplete = kwargs.get(r'allowincomplete', False)

    def getancestors(self, name, node, known=None):
        """Returns as many ancestors as we're aware of.

        return value: {
           node: (p1, p2, linknode, copyfrom),
           ...
        }
        """
        if known is None:
            known = set()
        if node in known:
            return []

        ancestors = {}
        def traverse(curname, curnode):
            # TODO: this algorithm has the potential to traverse parts of
            # history twice. Ex: with A->B->C->F and A->B->D->F, both D and C
            # may be queued as missing, then B and A are traversed for both.
            queue = [(curname, curnode)]
            missing = []
            seen = set()
            while queue:
                name, node = queue.pop()
                if (name, node) in seen:
                    continue
                seen.add((name, node))
                value = ancestors.get(node)
                if not value:
                    missing.append((name, node))
                    continue
                p1, p2, linknode, copyfrom = value
                if p1 != nullid and p1 not in known:
                    queue.append((copyfrom or curname, p1))
                if p2 != nullid and p2 not in known:
                    queue.append((curname, p2))
            return missing

        missing = [(name, node)]
        while missing:
            curname, curnode = missing.pop()
            try:
                ancestors.update(self._getpartialancestors(curname, curnode,
                                                           known=known))
                newmissing = traverse(curname, curnode)
                missing.extend(newmissing)
            except KeyError:
                # If we allow incomplete histories, don't throw.
                if not self.allowincomplete:
                    raise
                # If the requested name+node doesn't exist, always throw.
                if (curname, curnode) == (name, node):
                    raise

        # TODO: ancestors should probably be (name, node) -> (value)
        return ancestors

    @basestore.baseunionstore.retriable
    def _getpartialancestors(self, name, node, known=None):
        for store in self.stores:
            try:
                return store.getancestors(name, node, known=known)
            except KeyError:
                pass

        raise KeyError((name, hex(node)))

    @basestore.baseunionstore.retriable
    def getnodeinfo(self, name, node):
        for store in self.stores:
            try:
                return store.getnodeinfo(name, node)
            except KeyError:
                pass

        raise KeyError((name, hex(node)))

    def add(self, name, node, data):
        raise RuntimeError("cannot add content only to remotefilelog "
                           "contentstore")

    def getmissing(self, keys):
        missing = keys
        for store in self.stores:
            if missing:
                missing = store.getmissing(missing)
        return missing

    def markledger(self, ledger, options=None):
        for store in self.stores:
            store.markledger(ledger, options)

    def getmetrics(self):
        metrics = [s.getmetrics() for s in self.stores]
        return shallowutil.sumdicts(*metrics)

class remotefilelogmetadatastore(basestore.basestore):
    def getancestors(self, name, node, known=None):
        """Returns as many ancestors as we're aware of.

        return value: {
           node: (p1, p2, linknode, copyfrom),
           ...
        }
        """
        data = self._getdata(name, node)
        ancestors = shallowutil.ancestormap(data)
        return ancestors

    def getnodeinfo(self, name, node):
        return self.getancestors(name, node)[node]

    def add(self, name, node, parents, linknode):
        raise RuntimeError("cannot add metadata only to remotefilelog "
                           "metadatastore")

class remotemetadatastore(object):
    def __init__(self, ui, fileservice, shared):
        self._fileservice = fileservice
        self._shared = shared

    def getancestors(self, name, node, known=None):
        self._fileservice.prefetch([(name, hex(node))], force=True,
                                   fetchdata=False, fetchhistory=True)
        return self._shared.getancestors(name, node, known=known)

    def getnodeinfo(self, name, node):
        return self.getancestors(name, node)[node]

    def add(self, name, node, data):
        raise RuntimeError("cannot add to a remote store")

    def getmissing(self, keys):
        return keys

    def markledger(self, ledger, options=None):
        pass