httppeer: always add x-hg* headers to Vary header Before, we manually updated the Vary header value for each header contributing to it. All X-Hg* headers are reserved for the Mercurial protocol and could have caching implications. So it makes sense to always add these headers to Vary. A test revealed that X-HgArgs-Post wasn't being added to Vary. This is only sent on POST requests. POST requests generally aren't cacheable. However, it is possible if the server sends the appropriate headers. Mercurial shouldn't be sending those headers. But let's not take any chances. Differential Revision: https://phab.mercurial-scm.org/D3240

httppeer: don't accept very old media types (BC) Versions of Mercurial older than 1.0 emitted the text/plain and application/hg-changegroup media types in response to wire protocol commands. Way back in 8760d0c83b9b in 2005, the code validating these media types was added, presumably for backwards compatibility. 0b245edec124 a short time before that commit changed things from text/plain and application/hg-changegroup to application/mercurial-0.1 and application/hg-0.1. 8760d0c83b9b seemed to indicate ("for now") that the BC compatibility was temporary. But that code has lived until this day. It has been more than 10 years and nobody should be running pre 1.0 servers. Pretty much the only risk to this is if there's a server somewhere advertising the old media types or server software is interfering and not letting Mercurial send the proper Content-Type header. I think the chances are rare. The wire protocol docs were created (by me) from reading existing code. So the deletions don't constitute a spec change as much as reflecting the reality of how things have been for years. .. bc:: The HTTP client no longer accepts text/plain and application/hg-changegroup Content-Type values as a valid Mercurial command response. These should only be encountered on pre 1.0 Mercurial servers. Differential Revision: https://phab.mercurial-scm.org/D3239

httppeer: allow opener to be passed to makepeer() This allows us to use makepeer() in `hg debugwireproto`. Differential Revision: https://phab.mercurial-scm.org/D3238

httppeer: perform capabilities request in makepeer() Previously, we constructed an httppeer then always ran _fetchcaps() to issue the capabilities command. We want to issue the capabilities command before constructing a peer instance so we can construct an appropriate peer instance depending on the capabilities result. With the code for making and sending requests moved out of httppeer, it is now possible to send command requests without an httppeer. This commit creates a new function for making the capabilities request and calls it as part of makepeer(). This code should be functionality equivalent to what existed before. Differential Revision: https://phab.mercurial-scm.org/D3237

httppeer: extract common response handling into own function This allows the common redirect detection, content type validation, and decompression wrapping to be usable outside of httppeer instances. Differential Revision: https://phab.mercurial-scm.org/D3236

httppeer: move error handling and response wrapping into sendrequest This is common for all HTTP requests. It should be part of sendrequest(). Differential Revision: https://phab.mercurial-scm.org/D3235

httppeer: extract code for creating a request into own function Some of this feels awkward, such as having to pass in a function to evaluate a capability. And this code is generally pretty difficult to read. I didn't want to perform too much refactoring as part of the code move since it would make review more difficult. Differential Revision: https://phab.mercurial-scm.org/D3234

httppeer: extract code for performing an HTTP request This is generic and doesn't need to live as a method of httppeer. Differential Revision: https://phab.mercurial-scm.org/D3233

httppeer: move requestbuilder defaults into makepeer() argument Upcoming commits will move the initial ?cmd=capabilities handshake request out of httppeer so the handshake can be performed before a peer instance is constructed. In order to do this, we'll need to refactor code for making HTTP requests. The type used to construct HTTP requests is configurable. If we'll be making HTTP requests outside of httppeer, we should be able to use a custom request builder. So move the definition of that type into makepeer(). Extensions can monkeypatch the function and override the argument value. Differential Revision: https://phab.mercurial-scm.org/D3232

wireproto: move version 2 command handlers to wireprotov2server This is relatively straightforward. As part of this, we introduced a local @wireprotocommand that wraps the main one and defines a v2 only policy by default. Because the hacky HTTPv2 peer isn't using capabilities response yet, we had to move some code around to force import of wireprotov2server so commands are registered. This is super hacky. But this code will go away once the HTTPv2 peer is using the capabilities response to derive permissions. Differential Revision: https://phab.mercurial-scm.org/D3231

wireproto: extract HTTP version 2 code to own module wireprotoserver has generic and version 1 server code. The wireproto module also has both version 1 and version 2 command implementations. Upcoming work I want to do will make it difficult for this code to live in the current locations. Plus, it kind of makes sense for the version 2 code to live in an isolated module. This commit copies the HTTPv2 bits from wireprotoserver into a new module. We do it as a file copy to preserve history. A future commit will be copying wire protocol commands into this module as well. But there is little history of that code, so it makes sense to take history for wireprotoserver. Differential Revision: https://phab.mercurial-scm.org/D3230

wireproto: client reactor support for receiving frames We can now feed received frames into the client reactor and it will validate their sanity, dispatch them appropriately. The hacky HTTP peer has been updated to use the new code. No existing tests changed, somewhat proving the code works as expected. Rudimentary unit tests for the new functionality have been implemented. Differential Revision: https://phab.mercurial-scm.org/D3224

wireproto: introduce a reactor for client-side state We have a nice state machine of sorts for reacting to server-side events. Now it is time to implement the client equivalent. We introduce a "clientreactor." It allows callers to request that commands be issued. It has multiple modes of operation to reflect what the underlying transport supports. e.g. for SSH, we can perform wire sends immediately but for HTTP we need to buffer sends until all command requests are received. In addition, SSH allows sending multiple requests as long as the connection is open. But HTTP/1.1 only allows sending request data once. For SSH, we'll have one reactor per connection. For HTTP, we'll have one reactor per HTTP request. But because code that calls wire protocol commands should not be aware of how the underlying transport works, this will all be abstracted away by the peer interface. Our crude HTTP peer has been updated to use the reactor instead of formulating frames directly. No behavior should have changed here and tests seem to confirm that. Basic unit tests for the reactor behavior have been added. Differential Revision: https://phab.mercurial-scm.org/D3223

tests: extract wire protocol framing tests to own file I was lazy when I put these in test-wireproto-serverreactor.py. Let's do it properly. Differential Revision: https://phab.mercurial-scm.org/D3222

wireproto: disallow commands handlers for multiple transport versions I think it will be more trouble than it is worth to code version 1 and version 2 command handlers to the same interface. It will feel awkward to shoehorn functionality into e.g. the version 1 protocol handler interface. This would likely constrain the ability for version 2 to evolve. Previous commits introduced a clean separation between command handlers for version 1 and version 2 transports. This commit reinforces that separation by dropping support for having a single command handler service both version 1 and version 2 transports. Differential Revision: https://phab.mercurial-scm.org/D3208

wireproto: make @wireprotocommand version 1 only by default For backwards compatibility reasons. We want extension provided commands to opt in to version 2 rather than get inherited automatically. This will facilitate a clean break between the protocols. As part of this, we duplicate some commands used in tests so there are different command handlers per transport. Differential Revision: https://phab.mercurial-scm.org/D3207

wireproto: only expose "getbundle" and "unbundle" to v1 transports These are the most complicated wire protocol commands. I don't want to deal with porting them just yet. Let's disable both of them on version 2 transports so we drive the final wedge between command handlers and start to evolve version 2 command handlers more. Differential Revision: https://phab.mercurial-scm.org/D3206

wireproto: port lookup to wire protocol v2 This is pretty straightforward. We don't yet handle errors because we don't have an error handling mechanism in place yet. I'm also tempted to fold this into `known`. We'll come back to this later. Differential Revision: https://phab.mercurial-scm.org/D3205

wireproto: port pushkey command to wire protocol version 2 It doesn't do output redirection yet. And I'd love to generally overhaul the pushkey protocol for wire protocol version 2. But this will be a bit of effort. Let's do it as a follow-up. Differential Revision: https://phab.mercurial-scm.org/D3204

wireproto: only expose "clonebundles" to version 1 transports This may make a comeback in wire protocol version 2. The feature definitely needs to be carried forward. But at this juncture, I'm flirting with the idea of implementing this via a "redirect" mechanism at the command response level itself rather than something that requires one-off client support for querying and handling. i.e. I want to make it so servers can say "fetch this first and then come back" and clients handle that automatically. This would not only support clone bundles, but would also support piece-meal "pull bundles." Whatever happens, we can deal with it down the road. Differential Revision: https://phab.mercurial-scm.org/D3203

wireproto: define and expose types of wire command arguments Exposing the set of argument names is cool. But with wire protocol version 2, we're using CBOR to transport arguments and this allows us to have typing for arguments. Typed arguments are much nicer because they will cut down on transfer overhead and processing overhead for decoding values. This commit teaches @wireprotocommand to accept a dictionary for arguments. The arguments registered for version 2 transports are canonically stored as dictionaries rather than a space-delimited string. It is an error to defined arguments with a dictionary for commands using version 1 transports. This reinforces my intent to fully decouple command handlers for version 2 transports. Differential Revision: https://phab.mercurial-scm.org/D3202

wireproto: only expose "stream_out" to version 1 transports I have plans to implement stream clone using a better mechanism than this existing command. Let's not carry it forward to wire protocol version 2. Differential Revision: https://phab.mercurial-scm.org/D3201

wireproto: implement capabilities for wire protocol v2 The capabilities mechanism for wire protocol version 2 represents a clean break from version 1. Instead of effectively exchanging a set of capabilities, we're exchanging a rich data structure. This data structure currently contains information about every available command, including its accepted arguments. It also contains information about supported compression formats. Exposing information about supported commands will allow clients to automatically generate bindings to the server. Clients will be able to do things like detect when they are attempting to run a command that isn't known to the server. Exposing the required permissions to run a command can be used by clients to determine if they have privileges to call a command before actually calling it. We could potentially even have clients send credentials preemptively without waiting for the server to deny the command request. Lots of potential here. The data returned by this command will likely evolve heavily. So we shouldn't bikeshed the implementation just yet. Differential Revision: https://phab.mercurial-scm.org/D3200

context: add deprecation warnings for deprecated types of changeids It's close to code freeze, and dropping support for repo['123'] and repo ['my-bookmark'] and repo['deadbeef'] is pretty dispruptive, so this just adds deprecation warnings so extensions can easily find the places they need to fix. Differential Revision: https://phab.mercurial-scm.org/D3197

revsymbol: stop delegating to repo.__getitem__ for unhandled symbols (API) The only remaining cases where we were delegating unhandled symbols to repo.__getitem__ should now be when the symbol could not be found. In that case we just delegated to repo.__getitem__ for the error message. Let's just copy the error message instead. If there were any cases where we got e.g. a binary nodeid or an integer revnum into revsymbol() (e.g. via repo.lookup()), we'd now start raising an exception instead. That is why this is marked (API). This affects one test case, but the new behavior seems better to me. I can't tell if the old behavior was desired or if the test was just there to document how it happened to work. Differential Revision: https://phab.mercurial-scm.org/D3196

context: handle partial nodeids in revsymbol() Similar reasoning as previous patches. Differential Revision: https://phab.mercurial-scm.org/D3195

context: handle namespaces in revsymbol() Similar reasoning as previous patches. Differential Revision: https://phab.mercurial-scm.org/D3194

scmutil: handle full hex nodeids in revsymbol() This is a bit unfortunate, but it enables moving other pieces out of changectx's constructor without affecting the order in which we look up things (e.g. hex nodeid before bookmark). We convert nodeid to revnum before calling repo.__getitem__, even though that will result in converting back to nodeid later. This is so we can handle the LookupError and attempt to interpret the string as something else (e.g. a bookmark). We also need to start handling WdirUnsupported now, since the full hex nodeid "ffff..." represents the working directory. The exception is raised by the revlog layer. Differential Revision: https://phab.mercurial-scm.org/D3193

context: handle stringified ints in revsymbol() This patch copies the handling of stringified ints from changectx's constructor. It then calls repo.__getitem__ with the int. Since that method only interprets integers as revnums the first thing it does, this will not be redoing any of the work already done. We leave the old code in place so we can later deprecate it instead of breaking extensions. Differential Revision: https://phab.mercurial-scm.org/D3146

py3: make sure we write bytes to file # skip-blame because just b'' prefix Differential Revision: https://phab.mercurial-scm.org/D3220

py3: workaround comparing NoneType and integers Comparing None with integers was fine in Python 2 but returns error in Python 3 which is nice. This patch replaces None with -1 where sorting is done and some related logic. Differential Revision: https://phab.mercurial-scm.org/D3219

py3: use sys.stdout instead of print in test-mq-qpush-fail.t Differential Revision: https://phab.mercurial-scm.org/D3218

py3: use '//' for integer division in tests/test-mq-qimport.t Differential Revision: https://phab.mercurial-scm.org/D3217

py3: use pycompat.byterepr() instead of repr() Differential Revision: https://phab.mercurial-scm.org/D3216

py3: use bytes instead of str in isinstance() We deal with bytes internally and things should be bytes Differential Revision: https://phab.mercurial-scm.org/D3215

py3: use .startswith() instead of bytes[0] Differential Revision: https://phab.mercurial-scm.org/D3214

py3: add b'' prefixes in tests/test-ui-config.py # skip-blame because just b'' prefixes Differential Revision: https://phab.mercurial-scm.org/D3221

lfs: infer the blob store URL from paths.default If `lfs.url` is specified, it takes precedence. However, now that we support serving blobs via hgweb, we shouldn't *require* this setting. Less configuration is better (things will work out of the box once this is sorted out), and git has similar functionality. This is not a complete solution- it isn't able to infer the blob store from an explicitly supplied path, and it should consider `paths.default-push` for push. The pull solution for that is a bit hacky, and this alone is an improvement for the vast majority of cases. Even though there are only a handful of references to the saved remote store, the location of them makes things complicated. 1) downloading files on demand in the revlog flag processor 2) copying to readonlyvfs with bundlerepo 3) downloading in the file prefetch hook 4) the canupload()/skipdownload() checks 5) uploading blobs Since revlog doesn't have a repo or ui reference, we can't avoid creating a remote store when the extension is loaded. While the long term goal is to make sure the prefetch hook is invoked early for every command for efficiency, this handling in the flag processor is needed as a last ditch fetch. In order to support the clone command, the remote store needs to be created later than when the extension loads, since `paths.default` isn't set until just before the files are checked out. Therefore, this patch changes the prefetch hook to ignore the saved reference, and build a new one. The canupload()/skipdownload() checks simply check if the stored instance is a `_nullremote`. Since this can only be set via `lfs.url` (which is reflected in the saved reference), checking only the instance created when the extension loaded is fine. The blob uploading function is called from several places: 1) a prepush hook 2) when writing a new bundle 3) from infinitepush The prepush hook gets an exchange.pushop, so it has a path to where the push is going. The bundle writer and infinitepush don't. Further, bundle creation for things like strip and amend are causing blobs to be uploaded. This seems wrong, but I don't want to side track this sorting that out, so punt on trying to handle explicit push paths or `paths.default-push`. I also think that sending blobs to a remote store when pushing to a local repo is wrong. This functionality predates the usercache, so perhaps that's the reason for it. I've got some patches floating around to stop sending blobs remotely in this case, and instead write directly to the other repo's blob store. But the tests for corruption handling weren't happy with this change, and I don't have time to rewrite them. So exclude filesystem based paths from this for now. I don't think there's much of a chance to implement `paths.remote:lfsurl` style configs, given how early these are resolved vs how late the remote store is created. But git has it, so I threw a TODO in there, in case anyone has ideas. I have no idea why this is now doing http auth twice when it wasn't before. I don't think the original blobstore's url is ever being used in these cases.

lfs: add the ability to disable the usercache While the usercache is important for real world uses, I've been tripped up more than a couple of times by it in tests- thinking a file was being downloaded, but it was simply linked from the local cache. The syntax for setting it is the same as for setting a null remote endpoint, and like that endpoint, is left undocumented. This may or may not be a useful feature in the real world (I'd expect any sane filesystem to support hardlinks at this point).

tests: stabilize test-pull-bundle.t for Windows See 594dd384803c.

hgweb: wrap {archives} with mappinglist No bare list of mappings should be put in a template mapping.

hgweb: forward archivelist() of hgweb to webutil self.configlist() is ui.configlist(untrusted=True), and url=None in templater effectively means 'url' is undefined.

hgweb: move archivelist() of hgwebdir to webutil

hgweb: drop archivespecs from requestcontext It's a constant.

hgweb: move archivespecs to webutil Move one step further from e38e7ea21987. Since 'archivespecs' is a constant shared with hgweb and hgwebdir, webutil is a better place to define it.

hgweb: wrap {labels} by hybridlist() This one is a simple list of strings, which can support map operation.

hgweb: wrap {pathdef} with mappinglist No bare list of mappings should be put in a template mapping.

hgwebdir: wrap {entries} with mappinggenerator No bare generator should be put in a template mapping.

eol: look up partial nodeid as partial nodeid Similar reasoning as the previous patch. For some reason the hook gives us a partial nodeid, so we need to resolve that to a full nodeid. Differential Revision: https://phab.mercurial-scm.org/D3159

histedit: look up partial nodeid as partial nodeid I'm about to remove support for repo[<partial hex nodeid>]. In the verify() method, we know that self.node is always a partial or full binary nodeid, so the most correct way to look up the revision is by using changelog._partialmatch(), so let's do that. (It's closer to the current code to do scmutil.revsymbol(), but that's less correct because it will match a bookmark or tag that happens to have the same prefix as the node.) Differential Revision: https://phab.mercurial-scm.org/D3158

histedit: drop unnecessary check for "self.node is not None" We are doing hex(self.node) just a few lines up, so it shouldn't be None. The only way it could be none is if it was reassigned in between. The only way that can happen is if the user had put a "ffff..." wdirhex revision in the histedit script. This code is much older than the "ffff..." identifier, so I'm confident it's not there to handle that case. I'll let someone else add proper checks for "ffff..." if they care enough. Differential Revision: https://phab.mercurial-scm.org/D3157

context: extract partial nodeid lookup method to scmutil We will add another caller soon, and there's a non-obvious reason to use the unfiltered repo that we don't want to copy across the code base. Differential Revision: https://phab.mercurial-scm.org/D3189

templatekw: fix return type of {succsandmarkers} (BC) A hybrid object represents a list/dict of values, but {succsandmarkers} returns a list of template mappings. This change means old-style list templates (e.g. "start_succsandmarkers") are no longer supported, but that should be okay since {succsandmarkers} is still experimental and relatively new.

formatter: remove template resources from nested items before generating JSON

templater: add public interface returning a set of resource keys The next patch depends on knownresourcekeys(), and there's no reason to keep availableresourcekeys() private.

formatter: make nested items somewhat readable in template output

templater: wrap result of '%' operation so it never looks like a thunk This fixes min/max()/json() of map result. Before, it was taken as a lazy byte string and stringified by evalfuncarg().

wireproto: support for pullbundles Pullbundles are similar to clonebundles, but served as normal inline bundle streams. They are almost transparent to the client -- the only visible effect is that the client might get less changes than what it asked for, i.e. not all requested head revisions are provided. The client announces support for the necessary retries with the partial-pull capability. After receiving a partial bundle, it updates the set of revisions shared with the server and drops all now-known heads from the request list. It will then rerun getbundle until no changes are received or all remote heads are present. Extend badserverext to support per-socket limit, i.e. don't assume that the same limits should be applied to all sockets. Differential Revision: https://phab.mercurial-scm.org/D1856

filelog: wrap revlog instead of inheriting it (API) The revlog base class exposes a ton of methods. Inheriting the revlog class for filelog will make it difficult to expose a clean interface. There will be abstraction violations. This commit breaks the inheritance of revlog by the filelog class. Filelog instances now contain a reference to a revlog instance. Various properties and methods are now proxied to that instance. There is precedence for doing this: manifestlog does something similar. Although, manifestlog has a cleaner interface than filelog. We'll get there with filelog... The new filelog class exposes a handful of extra properties and methods that aren't part of the declared filelog interface. Every extra item was added in order to get a test to pass. The set of tests that failed without these extra proxies has significant overlap with the set of tests that don't work with the simple store repo. There should be no surprise there. Hopefully the hardest part about this commit to review are the changes to bundlerepo and unionrepo. Both repository types define a custom revlog or revlog-like class and then have a custom filelog that inherits from both filelog and their custom revlog. This code has been changed so the filelog types don't inherit from revlog. Instead, they replace the revlog instance on the created filelog. This is super hacky. I plan to fix this in a future commit by parameterizing filelog.__init__. Because Python function call overhead is a thing, this change could impact performance by introducing a nearly empty proxy function for various methods and properties. I would gladly measure the performance impact of it, but I'm not sure what operations have tight loops over filelog attribute lookups or function calls. I know some of the DAG traversal code can be sensitive about the performance of e.g. parentrevs(). However, many of these functions are implemented on the revlog class and therefore have direct access to self.parentrevs() and aren't going through a proxy. .. api:: filelog.filelog is now a standalone class and doesn't inherit from revlog. Instead, it wraps a revlog instance at self._revlog. This change was made in an attempt to formalize storage APIs and prevent revlog implementation details leaking through to callers. Differential Revision: https://phab.mercurial-scm.org/D3154

util: drop write_content_size=True This is now the default in python-zstandard 0.9. While we're here, also add a comment about the ability to drop frame magic to save space. Differential Revision: https://phab.mercurial-scm.org/D3199