Our project has a couple of remote subrepos in it, and their addresses have recently moved from http://host/path to http://other_host/path. How can one go back to a revision from, say, last month, where Mercurial thinks the subrepo can be found at http://host/path?
$ hg -v up -d 1/20/2012
Found revision 1091 from Fri Jan 20 10:22:29 2012 -0600
resolving manifests
abort: error: No connection could be made because the target machine actively refused it
$ hg --debug up -d 1/20/2012
Found revision 1091 from Fri Jan 20 10:22:29 2012 -0600
resolving manifests
<snip...>
subrepo merge 0f0f2b807811+ 0908d5249a6f 0f0f2b807811
subrepo external/our_remote_repo: both sides changed, merge with https://old_host/external/our_remote_repo:c66cf52ce1f240193190cec392d889618c09f22b:hg
using https://old_host/external/our_remote_repo
sending capabilities command
using auth.old_host.* for authentication
abort: error: No connection could be made because the target machine actively refused it
According to documentation you can use [subpaths] in hgrc to remap such urls.
This boils down to defining subrepositories source locations rewriting rules in the form of
<pattern> = <replacement>
where pattern is a regular expression matching the source and replacement is the replacement string used to rewrite it. Groups can be matched in pattern and referenced in replacements. For instance:
http://server/(.*)-hg/ = http://hg.server/\1/
rewrites
http://server/foo-hg/ into http://hg.server/foo/.
You can't do it. Mercurial returns in the state, which it had exactly at this revision (URL of subrepo, revision in subrepo)
Related
I am trying to clone this project with mercurial:
https://hg.mozilla.org/mozilla-central/file/tip/testing/geckodriver
I have downloaded TortoiseHg and selected clone repository.
It gives 404 error:
% hg clone --verbose https://hg.mozilla.org/mozilla-central/file/tip/testing/geckodriver "D:\Projects\gecko"
HTTP Error: 404 (Not Found)
[command returned code 255 Mon Oct 17 23:12:18 2022]
What is the correct path I should be trying to clone?
The URL used everywhere is just the URL of page in some Web-frontend for the some repository in some state, loosely connected to real URL of repository
I don't know the internal structure of Mozilla Central (they can use subrepos for separate projects or have one giant monorepo), but by inspecting changesets, which affected geckodriver dir in tree, I see modified files with path testing/geckodriver/ in names, based on which I assume that the project of interest is not an independent entity, but only a subtree in a large repository. Thus, because with HG you can't clone just part of tree (contrary to SVN, f.e.), you, maybe, have to clone the whole Mozilla Central repository in worst case
PS: hg clone -v https://hg.mozilla.org/mozilla-central/testing/geckodriver will not work, don't waste time on it
PPS: If I grok docs correctly (NB: it's not guaranteed!!!) you can enable (still experimental, but bundled for years) sparse extension, prepare sparse-config and try to clone needed subtree testing/geckodriver (it will be not dumb hg clone, but hg init, hack-hack hgrc, hg pull https://hg.mozilla.org/mozilla-central in order to have limiting sparse-profile in game on getting sources)
We discovered that our Hg repository got corrupted several weeks ago. This corruption seems to have propagated: all clones (the central repo and the user repos) are corrupted, pretty badly, in the same way. I think we could have prevented it if we did verification at that time.
Is there some Hg setting that would cause verification on each push, and prevent push in case of verification failure? I know I could implement it as a hook in Python, but is there maybe a simpler solution?
Is it also possible to do the opposite: make sure the remote repository is verified before pulling?
FWIW, I am on Windows 10 and we are using TortoiseHg.
Update: I've tried creating hooks as suggested by Jordi. Hg now hangs waiting for locks. Here is what I see:
c:\Users\username\test-hook>hg init
c:\Users\username\test-hook>cd ..
c:\Users\username>hg clone test-hook test-hook-clone
updating to branch default
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
# At this point I edited clone repository settings to include
# [hooks]
# preoutgoing = hg verify
#
# Then I created a test.txt file and "added" it via TortoiseHg context menu.
c:\Users\username\test-hook-clone>hg commit
c:\Users\username\test-hook-clone>hg status
c:\Users\username\test-hook-clone>hg outgoing
comparing with c:\Users\username\test-hook
searching for changes
changeset: 0:a61d33af6cdb
tag: tip
user: username
date: Mon May 06 20:32:54 2019 +0200
summary: test file added
c:\Users\username\test-hook-clone>hg push -verbose
pushing to c:\Users\username\test-hook
searching for changes
running hook preoutgoing: hg verify
waiting for lock on repository c:\Users\username\test-hook-clone held by process '16840' on host 'LT407233'
To answer your question, the hook does not have to be written in Python. In the appropriate server's hgrc (either at the repository level or the system level), simply set
[hooks]
preoutgoing = hg verify
preincoming = hg verify
This may significantly slow down all pull and push operations, but perhaps you are willing to sacrifice speed for correctness.
This will result in output like this when a client tries to pull from a corrupt repo:
$ hg clone http://localhost:9000 sample-repo
requesting all changes
remote: abort: preoutgoing hook exited with status 1
and in your server logs, you should see output similar to
127.0.0.1 - - [18/Apr/2019 12:41:09] "GET /?cmd=batch HTTP/1.1" 200 - x-hgarg-1:cmds=lheads+%3Bknown+nodes%3D x-hgproto-1:0.1 0.2 comp=zstd,zlib,none,bzip2 partial-pull
checking changesets
checking manifests
crosschecking files in changesets and manifests
checking files
a#0: broken revlog! (index data/a.i is corrupted)
warning: orphan data file 'data/a.i'
checked 2 changesets with 1 changes to 2 files
1 warnings encountered!
1 integrity errors encountered!
(first damaged changeset appears to be 0)
You can enable a server side option to perform more validation all incoming content, just set the server.validate=yes option on your server.
The simplest way is to enable it in your server global hgrc or in the repository .hg/hgrc file by adding the following two lines:
[server]
validate = yes
This is a server option but you can also use it on the client. It should validate pull too.
(By the way, what kind of corruption are you seeing?)
I have a mercurial repository with a clone on several computers. In particular, I have a Ubuntu computer where the repository resides in ~/.vim and a Windows computer with a related repository in C:\Users\ben\vimfiles.
This repository contains subrepositories, some of them git subrepositories. For example, from .hgsub:
pack/thirdparty/start/signify = [git]pack/thirdparty/start/signify
I wanted to merge changes made in each repository. So I cloned the repository from my Ubuntu computer to a USB stick, plugged it into my Windows computer, and pulled from the repository on the Windows computer into the clone on the USB stick. So far, so good.
Now I go to merge, or even just update to any of the versions coming from the Windows computer. I got an error like this:
pulling subrepo pack/thirdparty/start/signify from /home/ben/.vim/pack
/thirdparty/start/signify
fatal: 'C:/Program Files/Git/home/ben/.vim/pack/thirdparty/start/signify' does not appear to be a git repository
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.
abort: git fetch error 128 in pack/thirdparty/start/signify (in subrepo pack/thirdparty/start/signify)
[command returned code 255 Thu Dec 13 11:24:15 2018]
After a bunch of digging, I eventually solved the issue by manually editing the .git/config file in every single git subrepo to change the origin path from /home/ben/.vim to C:/Users/ben/vimfiles.
What's the correct way to do this using only Mercurial commands or configurations? Using TortoiseHg, when I update (but not merge!) I can select any of the saved paths to pull subrepos from. This seems to manually override the paths.default configuration for the pull operation. But, that setting seems to be ignored by the git subrepos.
From the mercurial's documentation on subrepositories (accessible on command line through hg help subrepos):
Subrepositories
...
Remapping Subrepositories Sources
A subrepository source location may change during a project life,
invalidating references stored in the parent repository history. To
fix this, rewriting rules can be defined in parent repository hgrc
file or in Mercurial configuration. See the [subpaths] section in
hgrc(5) for more details.
Documentation on subpaths:
subpaths
Subrepository source URLs can go stale if a remote server changes name
or becomes temporarily unavailable. This section lets you define
rewrite rules of the form:
<pattern> = <replacement>
where pattern is a regular expression matching a subrepository source
URL and replacement is the replacement string used to rewrite it.
Groups can be matched in pattern and referenced in replacements. For
instance:
http://server/(.*)-hg/ = http://hg.server/\1/
rewrites http://server/foo-hg/ into http://hg.server/foo/.
Relative subrepository paths are first made absolute, and the rewrite
rules are then applied on the full (absolute) path. If pattern doesn't
match the full path, an attempt is made to apply it on the relative
path alone. The rules are applied in definition order.
I have a repo on a network drive (served by Windows server), with local repos pushing/pulling to it on the various machines I work on.
I just dealt with this problem, and solved it by cloning the repo from the network drive to a local disk, pushing, then cloning it back again. The machine from which I did this had not problem pushing further changes after this.
Now I just tried pushing from my laptop, and this happens:
% hg --debug push "Z:\[main repo]"
pushing to Z:\[main repo]
query 1; heads
searching for changes
all remote heads known locally
listing keys for "bookmarks"
2 changesets found
list of changesets:
2ed25c8975482734e3b9eed828573fd711d26fd8
19a424c011ffd0c887cf1d54ed0b537a6c1af714
adding changesets
add changeset 2ed25c897548
add changeset 19a424c011ff
adding manifests
adding file changes
adding GEM.py revisions
transaction abort!
rollback completed
abort: No usable temporary file name found
[command returned code 255 Thu Mar 09 18:51:11 2017]
The only info pertaining to this error message I have found so far is this, and I definitely have no files named con.*in my project. There are several named con*.py but they have never been a problem, and both the laptop and my workstation are running Windows 7, and I've been working on this project for a few years now.
I have happily pushed from this laptop for over a year, and it was never a problem. I don't really have any good idea where to even start looking. Could it be connected to the fact that my workstation had the main repo opened at the same time? It was definitely not doing anything to it at the time.
Update:
I ran hg verify, and this is what it returns -- no problem as far as aI can tell
% hg --debug verify
repository uses revlog format 1
checking changesets
checking manifests
crosschecking files in changesets and manifests
checking files
73 files, 74 changesets, 226 total revisions
[command completed successfully Fri Mar 10 08:58:02 2017]
I had faced the same error as well.
I just ran tortoise hg as as administrator and that fixed it for me
I don't have an answer yet but I would try the following:
Update to the latest mercurial version (4.1) and try again
Verify the repo integrity with hg verify
Although I understand it always worked as is, try to rename all the con.py files. The thing with CON is that it represents a device, I think it comes from DOS times :-)
If I understand correctly, you push to Z:[main repo] where Z: is a Windows share. Try to push to the same repo in another way, with ssh (requires some setup, yes)
Good luck, very bizarre problem :-/
I have noticed by mercurial repository expanding in size when ever I use repo B to pull changes from repo A.
It seems that TortoiseHG creates files like hg-bundle-r3e6uf.hg10un under .hg directory. These files are usually 1-2MB in size each, so nothing too big, but together they create a lot, and can be an annoyance when doing backups.
This does not seem to happen if I pull changes instantly without reviewing them, or if I use repo A to push changes to B.
These bundle files seem useless as they are not copied when cloning the repository B.
Also the cloned repo is almost half smaller without them, so it is like data in these files wasn't moved to other files either.
Is it possible to:
A) Avoid creating these bundles on pull. (Pushing is option only when I have access to both repos)
B) Use some command to cleanup .hg directory. (Cloning is not very elegant)
EDIT:
When I select 'Incoming' first bundle is created:
% hg --repository C:\temp\hg\testB incoming --quiet --bundle c:\docume~1\username\locals~1\temp\thg.hlngus\CtemphgtestA_iavzew.hg C:\temp\hg\testA
1:d806c8cb0355
2:e0e3b20d5cb2
3:4e803a7ecefc
[command completed successfully Fri Aug 02 09:59:12 2013]
and then 'Accept', the second bundle is created:
% hg --repository C:\temp\hg\testB pull --verbose c:\docume~1\username\locals~1\temp\thg.hlngus\CtemphgtestA_iavzew.hg
pulling from c:\docume~1\username\locals~1\temp\thg.hlngus\CtemphgtestA_iavzew.hg
searching for changes
all local heads known remotely
3 changesets found
adding changesets
adding manifests
adding file changes
added 3 changesets with 3 changes to 1 files
(run 'hg update' to get a working copy)
[command completed successfully Fri Aug 02 10:00:10 2013]
Where as using 'Pull' directly, no extra bundles are created:
% hg --repository C:\temp\hg\testB pull --verbose C:\temp\hg\testA
pulling from C:\temp\hg\testA
searching for changes
all local heads known remotely
3 changesets found
adding changesets
adding manifests
adding file changes
added 3 changesets with 3 changes to 1 files
(run 'hg update' to get a working copy)
[command completed successfully Fri Aug 02 10:01:52 2013]
It seems this is TortoiseHg specific issue. Solution is either use push or use pull directly from command line to avoid extra bundles. Only (safe) way to cleanup seems to be the repository cloning.