[PATCH 11 of 11 RFC] url: refactor util.drop_scheme() and hg.localpath() into url.localpath()

Brodie Rao brodie at bitheap.org
Sat Mar 26 01:29:46 CDT 2011


# HG changeset patch
# User Brodie Rao <brodie at bitheap.org>
# Date 1301119907 25200
# Node ID 0d6a6ff7d3c7703672f40ed994b721ecd51e0222
# Parent  79716c21ef9a7816d71fc8da1b6a3dd051d507bb
url: refactor util.drop_scheme() and hg.localpath() into url.localpath()

This replaces util.drop_scheme() with url.localpath(), using url.url for
parsing instead of doing it on its own. The function is moved from
util to url to avoid an import cycle.

hg.localpath() is removed in favor of using url.localpath(). This
provides more consistent behavior between "hg clone" and other
commands.

This change has a side effect of making URLs like bundle://../foo not
correspond to relative paths with "hg bundle". This is because ".." is
considered to be the host for the URL. The URL should be bundle:../foo
to be relative.

Invalid URLs like the one above never worked with push, in, out, etc.,
so this makes URL handling more consistent.

Comparison of old and new behaviors:

URL                      drop_scheme()   hg.localpath()    url.localpath()
===                      =============   ==============    ===============
file://foo/foo           /foo            foo/foo           /foo
file://localhost:80/foo  /foo            localhost:80/foo  /foo
file://localhost:/foo    /foo            localhost:/foo    /foo
file://localhost/foo     /foo            /foo              /foo
file:///foo              /foo            /foo              /foo
file://foo               (empty string)  foo               /
file:/foo                /foo            /foo              /foo
file:foo                 foo             foo               foo
/foo                     /foo            /foo              /foo
file:///C:/foo           /C:/foo         /C:/foo           /C:/foo
file://C:/foo            /foo            C:/foo            /foo
file://D:/foo            /foo            D:/foo            /foo

On Windows:

URL                      drop_scheme()   hg.localpath()    url.localpath()
===                      =============   ==============    ===============
file://localhost:80/foo  C:/foo          localhost:80/foo  /foo
file://localhost:/foo    C:/foo          localhost:/foo    /foo
file://localhost/foo     C:/foo          C:/foo            /foo
file:///foo              C:/foo          C:/foo            /foo
file://foo               (empty string)  foo               /
file:/foo                /foo            /foo              /foo
file:foo                 foo             foo               foo
/foo                     /foo            /foo              /foo
file:///C:/foo           C:/C:/foo       /C:/foo           C:/foo
file:///D:/foo           C:/D:/foo       /D:/foo           D:/foo
file://C:/foo            C:/foo          C:/foo            C:/foo
file://D:/foo            C:/foo          D:/foo            D:/foo

The above behavior now also applies to the bundle scheme.

For more information about file:// URL handling, see:
http://www-archive.mozilla.org/quality/networking/testing/filetests.html

Related issues:

- issue1153: File URIs aren't handled correctly in windows

  This patch should preserve the fix implemented in
  2770d03ae49f. However, it goes a step further and "promotes"
  Windows-style drive letters from being interpreted as host names to
  being part of the path.

- issue2154: Cannot escape '#' in Mercurial URLs (#1172 in THG)

  The fragment is still interpreted as a revision or a branch, even in
  paths to bundles. This patch does not implement percent encoding
  handling.

diff --git a/mercurial/bundlerepo.py b/mercurial/bundlerepo.py
--- a/mercurial/bundlerepo.py
+++ b/mercurial/bundlerepo.py
@@ -15,7 +15,7 @@ from node import nullid
 from i18n import _
 import os, struct, tempfile, shutil
 import changegroup, util, mdiff, discovery
-import localrepo, changelog, manifest, filelog, revlog, error
+import localrepo, changelog, manifest, filelog, revlog, error, url
 
 class bundlerevlog(revlog.revlog):
     def __init__(self, opener, indexfile, bundle,
@@ -274,9 +274,9 @@ def instance(ui, path, create):
             cwd = os.path.join(cwd,'')
             if parentpath.startswith(cwd):
                 parentpath = parentpath[len(cwd):]
-    path = util.drop_scheme('file', path)
-    if path.startswith('bundle:'):
-        path = util.drop_scheme('bundle', path)
+    u = url.url(path)
+    path = u.localpath()
+    if u.scheme == 'bundle':
         s = path.split("+", 1)
         if len(s) == 1:
             repopath, bundlename = parentpath, s[0]
diff --git a/mercurial/hg.py b/mercurial/hg.py
--- a/mercurial/hg.py
+++ b/mercurial/hg.py
@@ -17,7 +17,7 @@ import verify as verifymod
 import errno, os, shutil
 
 def _local(path):
-    path = util.expandpath(util.drop_scheme('file', path))
+    path = util.expandpath(url.localpath(path))
     return (os.path.isfile(path) and bundlerepo or localrepo)
 
 def addbranchrevs(lrepo, repo, branches, revs):
@@ -102,15 +102,6 @@ def defaultdest(source):
     '''return default destination of clone if none is given'''
     return os.path.basename(os.path.normpath(source))
 
-def localpath(path):
-    if path.startswith('file://localhost/'):
-        return path[16:]
-    if path.startswith('file://'):
-        return path[7:]
-    if path.startswith('file:'):
-        return path[5:]
-    return path
-
 def share(ui, source, dest=None, update=True):
     '''create a shared repository'''
 
@@ -230,8 +221,8 @@ def clone(ui, source, dest=None, pull=Fa
     else:
         dest = ui.expandpath(dest)
 
-    dest = localpath(dest)
-    source = localpath(source)
+    dest = url.localpath(dest)
+    source = url.localpath(source)
 
     if os.path.exists(dest):
         if not os.path.isdir(dest):
@@ -257,7 +248,7 @@ def clone(ui, source, dest=None, pull=Fa
         abspath = origsource
         copy = False
         if src_repo.cancopy() and islocal(dest):
-            abspath = os.path.abspath(util.drop_scheme('file', origsource))
+            abspath = os.path.abspath(url.localpath(origsource))
             copy = not pull and not rev
 
         if copy:
diff --git a/mercurial/localrepo.py b/mercurial/localrepo.py
--- a/mercurial/localrepo.py
+++ b/mercurial/localrepo.py
@@ -1949,7 +1949,7 @@ def aftertrans(files):
     return a
 
 def instance(ui, path, create):
-    return localrepository(ui, util.drop_scheme('file', path), create)
+    return localrepository(ui, urlmod.localpath(path), create)
 
 def islocal(path):
     return True
diff --git a/mercurial/url.py b/mercurial/url.py
--- a/mercurial/url.py
+++ b/mercurial/url.py
@@ -63,6 +63,8 @@ class url(object):
         self.scheme = self.user = self.passwd = self.host = None
         self.port = self.path = self.query = self.fragment = None
         self._localpath = True
+        self._hostport = ''
+        self._origpath = path
 
         if not path.startswith('/') and ':' in path:
             parts = path.split(':', 1)
@@ -115,6 +117,7 @@ class url(object):
             # Don't split on colons in IPv6 addresses without ports
             if (self.host and ':' in self.host and
                 not (self.host.startswith('[') and self.host.endswith(']'))):
+                self._hostport = self.host
                 self.host, self.port = self.host.rsplit(':', 1)
                 if not self.host:
                     self.host = None
@@ -201,9 +204,31 @@ class url(object):
         return (s, (None, (str(self), self.host),
                     self.user, self.passwd or ''))
 
+    def localpath(self):
+        if self.scheme == 'file' or self.scheme == 'bundle':
+            path = self.path or '/'
+            # On Windows, we need to promote hosts containing drive
+            # letters to paths with drive letters.
+            if (os.name == 'nt' and len(self._hostport) == 2 and
+                self._hostport[0].isalpha() and self._hostport[1] == ':'):
+                path = self._hostport + '/' + self.path
+            elif self.host is not None and self.path:
+                path = '/' + path
+            # We also need to handle the case of file:///C:/, which
+            # should return C:/, not /C:/.
+            elif (os.name == 'nt' and path.startswith('/') and
+                  len(path) > 2 and path[1].isalpha() and path[2] == ':'):
+                # Strip leading slash from paths with drive names
+                return path[1:]
+            return path
+        return self._origpath
+
 def has_scheme(path):
     return bool(url(path).scheme)
 
+def localpath(path):
+    return url(path, parse_query=False, parse_fragment=False).localpath()
+
 def hidepassword(u):
     '''hide user credential in a url string'''
     u = url(u)
diff --git a/mercurial/util.py b/mercurial/util.py
--- a/mercurial/util.py
+++ b/mercurial/util.py
@@ -1378,26 +1378,6 @@ def bytecount(nbytes):
             return format % (nbytes / float(divisor))
     return units[-1][2] % nbytes
 
-def drop_scheme(scheme, path):
-    sc = scheme + ':'
-    if path.startswith(sc):
-        path = path[len(sc):]
-        if path.startswith('//'):
-            if scheme == 'file':
-                i = path.find('/', 2)
-                if i == -1:
-                    return ''
-                # On Windows, absolute paths are rooted at the current drive
-                # root. On POSIX they are rooted at the file system root.
-                if os.name == 'nt':
-                    droot = os.path.splitdrive(os.getcwd())[0] + '/'
-                    path = os.path.join(droot, path[i + 1:])
-                else:
-                    path = path[i:]
-            else:
-                path = path[2:]
-    return path
-
 def uirepr(s):
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')
diff --git a/tests/test-bundle.t b/tests/test-bundle.t
--- a/tests/test-bundle.t
+++ b/tests/test-bundle.t
@@ -139,7 +139,7 @@ Log -R full.hg in fresh empty
   $ rm -r empty
   $ hg init empty
   $ cd empty
-  $ hg -R bundle://../full.hg log
+  $ hg -R bundle:../full.hg log
   changeset:   8:aa35859c02ea
   tag:         tip
   parent:      3:eebf5a27f8ca
@@ -196,6 +196,12 @@ Make sure bundlerepo doesn't leak tempfi
   requires
   store
 
+Pull invalid relative URL
+
+  $ hg pull bundle://../full.hg
+  abort: No such file or directory: /full.hg
+  [255]
+
 Pull ../full.hg into empty (with hook)
 
   $ echo '[hooks]' >> .hg/hgrc
@@ -203,10 +209,10 @@ Pull ../full.hg into empty (with hook)
 
 doesn't work (yet ?)
 
-hg -R bundle://../full.hg verify
+hg -R bundle:../full.hg verify
 
-  $ hg pull bundle://../full.hg
-  pulling from bundle://../full.hg
+  $ hg pull bundle:../full.hg
+  pulling from bundle:../full.hg
   requesting all changes
   adding changesets
   adding manifests
@@ -257,7 +263,7 @@ Create partial clones
 
 Log -R full.hg in partial
 
-  $ hg -R bundle://../full.hg log
+  $ hg -R bundle:../full.hg log
   changeset:   8:aa35859c02ea
   tag:         tip
   parent:      3:eebf5a27f8ca
@@ -309,8 +315,8 @@ Log -R full.hg in partial
 
 Incoming full.hg in partial
 
-  $ hg incoming bundle://../full.hg
-  comparing with bundle://../full.hg
+  $ hg incoming bundle:../full.hg
+  comparing with bundle:../full.hg
   searching for changes
   changeset:   4:095197eb4973
   parent:      0:f9ee2f85a263
@@ -343,7 +349,7 @@ Incoming full.hg in partial
 
 Outgoing -R full.hg vs partial2 in partial
 
-  $ hg -R bundle://../full.hg outgoing ../partial2
+  $ hg -R bundle:../full.hg outgoing ../partial2
   comparing with ../partial2
   searching for changes
   changeset:   4:095197eb4973
@@ -377,7 +383,7 @@ Outgoing -R full.hg vs partial2 in parti
 
 Outgoing -R does-not-exist.hg vs partial2 in partial
 
-  $ hg -R bundle://../does-not-exist.hg outgoing ../partial2
+  $ hg -R bundle:../does-not-exist.hg outgoing ../partial2
   abort: No such file or directory: ../does-not-exist.hg
   [255]
   $ cd ..
@@ -470,6 +476,22 @@ test for 540d1059c802
   
   $ cd ..
 
+test bundle with # in the filename (issue2154):
+
+  $ cp bundle.hg 'test#bundle.hg'
+  $ cd orig
+  $ hg incoming '../test#bundle.hg'
+  comparing with ../test
+  abort: unknown revision 'bundle.hg'!
+  [255]
+
+note that percent encoding is not handled:
+
+  $ hg incoming ../test%23bundle.hg
+  abort: repository ../test%23bundle.hg not found!
+  [255]
+  $ cd ..
+
 test for http://mercurial.selenic.com/bts/issue1144
 
 test that verify bundle does not traceback
diff --git a/tests/test-pull.t b/tests/test-pull.t
--- a/tests/test-pull.t
+++ b/tests/test-pull.t
@@ -71,6 +71,10 @@ Test 'file:' uri handling:
   abort: repository /test-doesnt-exist not found!
   [255]
 
+  $ hg pull -q file://../test
+  abort: repository /test not found!
+  [255]
+
   $ hg pull -q file:../test
 
 It's tricky to make file:// URLs working on every platform with


More information about the Mercurial-devel mailing list