#+title: Forge User and Developer Manual
:PREAMBLE:
#+author: Jonas Bernoulli
#+email: emacs.forge@jonas.bernoulli.dev
#+date: 2018-{{{year}}}

#+texinfo_dir_category: Emacs
#+texinfo_dir_title: Forge: (forge).
#+texinfo_dir_desc: Access Git Forges from Magit
#+subtitle: for version 0.6.2

#+setupfile: .orgconfig

Forge allows you to work with Git forges, currently Github and Gitlab,
from the comfort of Magit and Emacs.

#+texinfo: @noindent
This manual is for Forge version 0.6.2.

#+texinfo: @insertcopying
:END:
* Introduction

Forge allows you to work with Git forges, currently Github and Gitlab,
from the comfort of Magit and Emacs.

Forge fetches issues, pull-requests and other data using the forge's
API and stores the retrieved information in a local database.
Additionally it fetches pull-request references using Git.

* Initial Setup

Please first do the common setup below and then carefully follow the
instructions for your forge instance.  Once you have completed the
setup, you can start tracking repositories (see [[*Initial Pull]]).

If you run into difficulties during setup or the initial pull, then
please also see [[*How Forge Detection Works]] and [[info:ghub#Getting
Started]].

** Common Setup
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Loading Magit doesn't cause Forge to be loaded automatically.  Adding
something like this to your init file takes care of that:

#+begin_src emacs-lisp
  (with-eval-after-load 'magit
    (require 'forge))
#+end_src

#+texinfo: @noindent
Or if you use ~use-package~:

#+begin_src emacs-lisp
  (use-package forge
    :after magit)
#+end_src

By default Forge adds some bindings to Magit keymaps and menus, and
some sections to Magit buffers.  If you would like to prevent that,
you have to set ~forge-add-default-bindings~ and/or
~forge-add-default-sections~ to ~nil~, before ~magit~ (not just ~forge~) is
loaded.

** Setup for Github.com
*** Set your Username
:PROPERTIES:
:UNNUMBERED: notoc
:END:

First inform Forge about your https://github.com username:

#+begin_src shell
  git config --global github.user USERNAME
#+end_src

If you need to identify as another user in a particular repository,
then you have to set that variable locally:

#+begin_src shell
  cd /path/to/repo
  git config --local github.user USERNAME
#+end_src

*** Create and Store an Access Token
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Visit https://github.com/settings/tokens in a browser to generate
a new "classic" token using the ~repo~, ~user~ and ~read:org~ scopes.
Do not close the browser window just yet, because the token will
only be shown once.

The built-in Auth-Source ([[info:auth]]) package is used to store the
token generated in the previous step.  The ~auth-sources~ variable
controls how and where Auth-Source keeps its secrets.  The default
value is a list of three files: ~("~/.authinfo" "~/.authinfo.gpg"
"~/.netrc")~, but that can lead to confusing behavior, so you should
make sure that only one of these files exists, and then you should
also adjust the value of the variable to only ever use that file,
for example:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo"))
#+end_src

In ~~/.authinfo~ secrets are stored in plain text.  If you don't want
that, then you should use the encrypted ~~/.authinfo.gpg~ instead:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo.gpg"))
#+end_src

Make sure you put one of these forms in your init file *and* to evaluate
it in the current Emacs instance as well, by placing the cursor after
the final closing parenthesis and typing ~C-x C-e~ (~eval-last-sexp~).

Next add a line like the following to the chosen file:

#+begin_src example
  machine api.github.com login USERNAME^forge password TOKEN
#+end_src

- The value of ~machine~ must be ~api.github.com~.  Variations of this
  won't work.

- USERNAME must be the same as the value used for the ~github.user~ Git
  variable above.  You *must* append ~^forge~ to that, without any space
  in between.

- TOKEN is the token you generated earlier.

Finish by typing ~M-x auth-source-forget-all-cached RET~.  If you don't
do this, then Auth-Source may fail to look up the token.

** Setup for Another Github Instance

Before you setup a Github instance that is not https://github.com,
please set that up first.  The setup for https://github.com is easier
and if that works, but the setup for the other Github instance fails,
then we can tentatively narrow the issue down to the parts that differ
between https://github.com and other instances.

*** Tell Forge about the Instance
:PROPERTIES:
:UNNUMBERED: notoc
:END:

While Forge knows about https://github.com, it does not know about
your other Github instances.  Forge instances are configured using
the option ~forge-alist~ (also see its docstring).  The entry for
https://github.com in that variable looks like this:

#+begin_src emacs-lisp
  ("github.com"                       ; GITHOST
   "api.github.com"                   ; APIHOST
   "github.com"                       ; WEBHOST and INSTANCE-ID
    forge-github-repository)          ; CLASS
#+end_src

You have to add an entry for your instance.  For example, assuming
you company uses https://example.com, this might be correct:

#+begin_src emacs-lisp
  (push '("example.com"               ; GITHOST
          "api.example.com"           ; APIHOST
          "example.com"               ; WEBHOST and INSTANCE-ID
          forge-github-repository)    ; CLASS
        forge-alist)
#+end_src

Your company may use hostnames that follow a different format.  You
should be able to easily determine and verify GITHOST and WEBHOST,
but determining APIHOST is more difficult; you might have to ask a
coworker.  APIHOST could be something like ~api.example.com~, but it
could also be something like ~example.com/api~.

If the REST API's end point is ~/v3~ and the GraphQL API's end point is
~/graphql~, then use something like ~example.com/v3~ as APIHOST.  This is
a historic accident.  See https://github.com/magit/forge/issues/174.

We will use INSTANCE-ID (aka WEBHOST) and APIHOST below.

*** Set your Username
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Inform Forge about your username for the Github instance in question:

#+begin_src shell
  git config --global github.INSTANCE-ID.user USERNAME
#+end_src

So if INSTANCE-ID is ~example.com~ and USERNAME is ~tarsius~ then use:

#+begin_src shell
  git config --global github.example.com.user tarsius
#+end_src

*** Create and Store an Access Token
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Visit your forge in a browser.  Follow a link to "Settings", from
there to "Developer settings", from there to "Personal access tokens",
and finally to "Tokens (classic)".  On that page generate a new token
using the ~repo~, ~user~ and ~read:org~ scopes.  Do not close the browser
window just yet, because the token will only be shown once.

The built-in Auth-Source ([[info:auth]]) package is used to store the
token generated in the previous step.  The ~auth-sources~ variable
controls how and where Auth-Source keeps its secrets.  The default
value is a list of three files: ~("~/.authinfo" "~/.authinfo.gpg"
"~/.netrc")~, but that can lead to confusing behavior, so you should
make sure that only one of these files exists, and then you should
also adjust the value of the variable to only ever use that file,
for example:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo"))
#+end_src

In ~~/.authinfo~ secrets are stored in plain text.  If you don't want
that, then you should use the encrypted ~~/.authinfo.gpg~ instead:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo.gpg"))
#+end_src

Make sure you put one of these forms in your init file *and* to evaluate
it in the current Emacs instance as well, by placing the cursor after
the final closing parenthesis and typing ~C-x C-e~ (~eval-last-sexp~).

Next add a line like the following to the chosen file:

#+begin_src example
  machine APIHOST login USERNAME^forge password TOKEN
#+end_src

- APIHOST must be the same as the second element of the entry we added
  to ~forge-alist~.  In the above example that would be ~api.example.com~.
  Do *not* instead use GITHOST or INSTANCE-ID (aka WEBHOST).

- USERNAME must be the same username you used above as the value of
  the Git variable.  You *must* append ~^forge~ to that, without any space
  in between.

- TOKEN is the token you generated earlier.

Finish by typing ~M-x auth-source-forget-all-cached RET~.  If you don't
do this, then Auth-Source may fail to look up the token.

** Setup for Gitlab.com
*** Set your Username
:PROPERTIES:
:UNNUMBERED: notoc
:END:

First inform Forge about your https://gitlab.com username:

#+begin_src shell
  git config --global gitlab.user USERNAME
#+end_src

If you need to identify as another user in a particular repository,
then you have to set that variable locally:

#+begin_src shell
  cd /path/to/repo
  git config --local gitlab.user USERNAME
#+end_src

*** Create and Store an Access Token
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Visit https://gitlab.com/-/user_settings/personal_access_tokens in a
browser to generate a new token using the ~api~, ~read_api~ and ~read_user~
scopes.  Do not close the browser window just yet, because the token
will only be shown once.

The built-in Auth-Source ([[info:auth]]) package is used to store the
token generated in the previous step.  The ~auth-sources~ variable
controls how and where Auth-Source keeps its secrets.  The default
value is a list of three files: ~("~/.authinfo" "~/.authinfo.gpg"
"~/.netrc")~, but that can lead to confusing behavior, so you should
make sure that only one of these files exists, and then you should
also adjust the value of the variable to only ever use that file,
for example:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo"))
#+end_src

In ~~/.authinfo~ secrets are stored in plain text.  If you don't want
that, then you should use the encrypted ~~/.authinfo.gpg~ instead:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo.gpg"))
#+end_src

Make sure you put one of these forms in your init file *and* to evaluate
it in the current Emacs instance as well, by placing the cursor after
the final closing parenthesis and typing ~C-x C-e~ (~eval-last-sexp~).

Next add a line like the following to the chosen file:

#+begin_src example
  machine gitlab.com login USERNAME^forge password TOKEN
#+end_src

- The value of ~machine~ should be ~gitlab.com~.  For historic reasons
  ~gitlab.com~ is also supported.

- Note that if you instead use the ~auth-source-pass~ backend, then you
  *must* use ~gitlab.com~ as ~host~.

- USERNAME must be the same as the value used for the ~gitlab.user~ Git
  variable above.  You *must* append ~^forge~ to that, without any space
  in between.

- TOKEN is the token you generated earlier.

Finish by typing ~M-x auth-source-forget-all-cached RET~.  If you don't
do this, then Auth-Source may fail to look up the token.

** Setup for Another Gitlab Instance

Before you setup a Gitlab instance that is not https://gitlab.com,
please set that up first.  The setup for https://gitlab.com is easier
and if that works, but the setup for the other Gitlab instance fails,
then we can tentatively narrow the issue down to the parts that differ
between https://gitlab.com and other instances.

*** Tell Forge about the Instance
:PROPERTIES:
:UNNUMBERED: notoc
:END:

While Forge knows about https://gitlab.com (and a few other well-known
instances, see its value) it has to be taught about other Gitlab
instances.  Forge instances are configured using the option
~forge-alist~ (also see its docstring).  The entry for
https://gitlab.com in that variable looks like this:

#+begin_src emacs-lisp
  ("gitlab.com"                       ; GITHOST
   "gitlab.com/api/v4"                ; APIHOST
   "gitlab.com"                       ; WEBHOST and INSTANCE-ID
   forge-gitlab-repository)           ; CLASS
#+end_src

For historic reasons, APIHOST actually has to be a host followed by
a path.

You have to add an entry for your instance.  For example, assuming you
company/organisation uses https://example.com, this might be correct:

#+begin_src emacs-lisp
  (push '("example.com"               ; GITHOST
          "example.com/api/v4"        ; APIHOST
          "example.com"               ; WEBHOST and INSTANCE-ID
          forge-gitlab-repository)    ; CLASS
        forge-alist)
#+end_src

Your company may use hostnames that follow a different format.  You
should be able to easily determine and verify GITHOST and WEBHOST,
but determining APIHOST is more difficult; you might have to ask a
colleague.

We will use INSTANCE-ID (aka WEBHOST) and APIHOST below.

*** Set your Username
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Inform Forge about your username for the Gitlab instance in question:

#+begin_src shell
  git config --global gitlab.INSTANCE-ID.user USERNAME
#+end_src

So if INSTANCE-ID is ~example.com~ and USERNAME is ~tarsius~ then use:

#+begin_src shell
  git config --global gitlab.example.com.user tarsius
#+end_src

*** Create and Store an Access Token
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Visit your forge in a browser.  Follow a link to "Preferences" and
from there to "Access Tokens".  On that page generate a new "Personal
access token" using the ~api~, ~read_api~ and ~read_user~ scopes.  Do not
close the browser window just yet, because the token will only be
shown once.

The built-in Auth-Source ([[info:auth]]) package is used to store the
token generated in the previous step.  The ~auth-sources~ variable
controls how and where Auth-Source keeps its secrets.  The default
value is a list of three files: ~("~/.authinfo" "~/.authinfo.gpg"
"~/.netrc")~, but that can lead to confusing behavior, so you should
make sure that only one of these files exists, and then you should
also adjust the value of the variable to only ever use that file,
for example:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo"))
#+end_src

In ~~/.authinfo~ secrets are stored in plain text.  If you don't want
that, then you should use the encrypted ~~/.authinfo.gpg~ instead:

#+begin_src emacs-lisp
  (setq auth-sources '("~/.authinfo.gpg"))
#+end_src

Make sure you put one of these forms in your init file *and* to evaluate
it in the current Emacs instance as well, by placing the cursor after
the final closing parenthesis and typing ~C-x C-e~ (~eval-last-sexp~).

Next add a line like the following to the chosen file:

#+begin_src example
  machine APIHOST login USERNAME^forge password TOKEN
#+end_src

- APIHOST must be either the same as the second element of the entry
  we added to ~forge-alist~, or that element with the path suffix
  removed.  For example, if the APIHOST, specified in an ~forge-alist~
  entry, is ~example.com/api/v4~, then you can use ~example.com/api/v4~ or
  ~example.com~.

  Historically only the former was supported, but that was a design
  mistake, which we are now stuck with, and using just ~example.com~ as
  APIHOST in ~~/.authsource~ is now recommended.  For the time being, in
  ~forge-alist~ you must still always use ~example.com/api/v4~ as APIHOST.

  If you instead use the ~auth-source-pass~ backend, then you *must* use
  ~example.com~ as the value of ~host~.  Using ~example.com/api/v4~ does not
  work with that backend.

- USERNAME must be the same username you used above as the value of
  the Git variable.  You *must* append ~^forge~ to that, without any space
  in between.

- TOKEN is the token you generated earlier.

Finish by typing ~M-x auth-source-forget-all-cached RET~.  If you don't
do this, then Auth-Source may fail to look up the token.

** Setup a Partially Supported Host

Forge currently only supports the Github and Gitlab APIs.

It does however partially support a few additional forge types (see
[[*Partially Supported Forges]]) and other lighter weight software used
to host Git repositories, which also provide a web interfaces (see
[[*Supported Semi-Forges]]).  Forge doesn't use the APIs of such forges,
but registering the host and adding repositories to the local database
at least enables the use of commands such as ~forge-browse~.

*** Tell Forge about the Instance
:PROPERTIES:
:UNNUMBERED: notoc
:END:

A few hosts, which use partially supported forge types, are available
out-of-the-box, because they have an entry in the default value of
option ~forge-alist~ (also see its docstring).  For example, the entry
for https://codeberg.org in that variable looks like this:

#+begin_src emacs-lisp
  ("codeberg.org"                     ; GITHOST
   "codeberg.org/api/v1"              ; APIHOST
   "codeberg.org"                     ; WEBHOST and INSTANCE-ID
   forge-gitea-repository)            ; CLASS
#+end_src

To be able to add repositories from a, so far, unknown forge instance
to your local database, you have to add an entry for that instance to
~forge-alist~.  For example, assuming you use another Gitea instance,
hosted at https://example.com, this might be correct:

#+begin_src emacs-lisp
  (push '("example.com"               ; GITHOST
          "example.com/api/v1"        ; APIHOST
          "example.com"               ; WEBHOST and INSTANCE-ID
          forge-gitea-repository)     ; CLASS
        forge-alist)
#+end_src

Look at ~forge-alist~ entries of other hosts using the same forge type
as the instance you are configuring, to see what format *might* be
appropriate.  You should be able to easily determine and verify
GITHOST and WEBHOST, but determining APIHOST is more difficult; you
might have to ask a colleague.  APIHOST could be something like
~example.com/api/vi~, but it could also be something like
~api.example.com~.

*** Add Support for Additional Forge Types
:PROPERTIES:
:UNNUMBERED: notoc
:END:

For each fully or partially supported forge type, Forge defines at
least a class.  The following example is taken from ~forge-semi.el~:

#+begin_src emacs-lisp
(defclass forge-cgit-repository (forge-noapi-repository)
  ((commit-url-format :initform "https://%h/%p.git/commit/?id=%r")
   (branch-url-format :initform "https://%h/%p.git/log/?h=%r")
   (remote-url-format :initform "https://%h/%p.git/about"))
  "Cgit from https://git.zx2c4.com/cgit/about.
Different hosts use different url schemata, so we need multiple
classes.  See their definitions in \"forge-semi.el\".")
#+end_src

Once you add a host using that class to ~forge-alist~ and then a
repository from that host to the local database, you will be able
to use commands such as ~forge-browse-branch~ (but not much more).

If you want to add a repository from another host, which happens to
use another software or another URL schemata, then you might have to
define an additional class first.  See ~forge-semi.el~ for simple
examples and grep for ~defclass forge-.*-repository~ for more complex
ones.

* Initial Pull

To start using Forge in a certain repository, visit the Magit status
buffer for that repository and type ~N / a~ (~forge-add-repository~).  You
are given a choice to pull all topics, all topics that were updated
after a certain date, or only individual topics.

Beside adding the repository to the database, this also adds a new
value to the Git variable ~remote.<remote>.fetch~, which causes all
pull-request refs (~+refs/pull/*/head:refs/pullreqs/*~ for Github)
to be fetched by Git.

Note that it is possible to use the same command to add any repository
from a supported forge to the database, without cloning the Git
repository first.

The initial fetch can take a while but most of the work is done
asynchronously.  Storing the information in the database is done
synchronously though, so there can be a noticeable hang at the end.
Subsequent fetches are much faster.

Fetching issues from Github is much faster than fetching from other
forges, because making a handful of GraphQL requests, is much faster
than making hundreds of REST requests.

* Getting Started

Much like Git stores information in a local repository and does not
require a constant internet connection, Forge retrieves additional
information using a forge's API and stores that in a local database.

Forge's equivalent of ~git clone~ is ~forge-add-repository~, which has to
be run, before most of Forges features become available in the local
clone of a Git repository.

- Key: N / a (forge-add-repository) ::

  This command guides the user through the process of adding a
  repository to the local database.

  Note that it is possible to add a repository to the local database,
  without pulling all the data, which is useful if you just want to
  create a single issue or pull-request in a repository, but are not
  interested in existing topics, e.g., because you do not regularly
  contribute to that repository.

  Also note that you can add a repository to the local database, even
  if no local Git clone exists.

Like with Git, you have to explicitly pull remote changes, at your
leisure, using ~forge-pull~.

- Key: f n (forge-pull) ::
- Key: N f f ::

  This command uses a forge's API to fetch topics and other
  information about the current repository, and stores the fetched
  information in the database.

  If the current repository isn't being tracked in the local database
  yet, then this command pivots to behave like ~forge-add-repository~.

Forge adds two additional sections to Magit's status buffer, which
list open and/or pending issues and pull-requests.  Typing ~RET~, while
the cursor is on a topic section, shows more information about that
topic in a separate buffer.  Typing ~RET~ on a topic list section, shows
that list in a separate buffer, where you can apply different filters.

The other main entry point to the functionality provided by Forge is
the ~forge-dispatch~ menu.

- Key: N (forge-dispatch) ::

  This prefix command is available in all Magit buffers and provides
  access to most of the available Forge commands.  See the following
  sections for information about the available commands.

* Lists and Menus

Topics are listed in two sections in Magit's status buffer, but can
also be listed in dedicated buffers.  Likewise individual topics can
be visited in separate buffers.  In both cases this can be done by
placing the cursor on the respective section in the status buffer and
typing ~RET~, or by invoking the appropriate command from Forge's main
menu, on ~N~ (~forge-dispatch~).

List commands and corresponding menu commands exist for topics,
notifications and repositories, but there isn't always an exclusive
mapping from menu to buffer.  The main menu (~forge-dispatch~), the
configuration menu (~forge-configure~), the menu which controls the
current topic or the topic at point (~forge-topic-menu~), and the
menu which controls the topics listed in the current buffer
(~forge-topics-menu~), are useful in more than one major mode.

All of these menus feature bindings to directly switch to the other
appropriate menus.  So it is enough to remember that ~N~ always brings
up the dispatch menu; you can always navigate to another menu from
there.

~C-c C-c~ brings up the most appropriate menu for the current buffer.
In Magit's status buffer the most appropriate menu is Magit's own
dispatch menu (~magit-dispatch~), so here the quickest way to invoke
Forge's dispatch menu is ~N~.  Even in Magit's status buffer, when the
cursor is an individual topic or on a topic list section, ~C-c C-c~
opens the respective menu (~forge-topics-menu~ or ~forge-topic-menu~).

The following sections describe most of the available menu and list
commands.  For ~forge-topic-menu~, see [[*Editing Topics]].

** Dispatch and configuration menus
:PROPERTIES:
:UNNUMBERED: notoc
:END:

- Key: N (forge-dispatch) ::

  This prefix menu command is available in all Magit buffers and
  provides access to most of the available Forge commands.  See the
  following sections for information about the available commands.

- Key: N m c (forge-configure) ::

  This command displays a menu used to configure the current
  repository and some global settings as well.

** Topic menu and list commands
:PROPERTIES:
:UNNUMBERED: notoc
:END:

- Key: N m f (forge-topics-menu) ::
- Key: C-c C-c [in topics list buffer/section] ::

  This command displays a menu used to control the list of topics
  displayed in the current buffer.

  Note that this command can not only be used in buffers dedicated
  to listing topics, but also in Magit's status buffer.

- Key: N l t (forge-list-topics) ::

  This command lists the current repository's issues in a separate
  buffer.  If the list buffer already exists, this command only
  ensures that all types of topics are listed.  If any other filters
  are in effect, they are left intact.

- Key: RET [on "Issues" status section] (forge-list-issues) ::

  This command lists the current repository's issues in a separate
  buffer.  If the list buffer already exists, this command limits the
  list to issues.  If any other filters are in effect, they are left
  intact.

- Key: RET [on "Pull requests" status section] (forge-list-pullreqs) ::

  This command lists the current repository's pull-requests in a
  separate buffer.  If the list buffer already exists, this command
  limits the list to pull-requests.  If any other filters are in
  effect, they are left intact.

- Key: N l g (forge-list-global-topics) ::

  This command lists topics across all tracked repository.  If the
  list buffer already exists, filters except for the type filter are
  left in effect.

- Command: forge-list-global-issues ::

  This command lists issues across all tracked repository.  If the
  list buffer already exists, filters except for the type filter are
  left in effect.

- Command: forge-list-global-pullreqs ::

  This command lists pull-requests across all tracked repository.  If
  the list buffer already exists, filters except for the type filter
  are in effect.

** Notification menu and list commands
:PROPERTIES:
:UNNUMBERED: notoc
:END:

- Key: N m n (forge-notifications-menu) ::
- Key: C-c C-c [in notifications list buffer] ::

  This command displays a menu used to control the list of
  notifications displayed in the current buffer.

- Key: N l n (forge-list-notifications) ::

  This command lists all notifications for all forges in a separate
  buffer.

** Repository menu and list commands
:PROPERTIES:
:UNNUMBERED: notoc
:END:

- Key: N m r (forge-repositories-menu) ::
- Key: C-c C-c [in repositories list buffer] ::

  This command displays a menu used to control the list of
  repositories displayed in the current buffer.

- Key: N l r (forge-list-repositories) ::

  This command lists all known repositories in a separate buffer.
  Here "known" means that an entry exists in the local database.

- Key: RET [on repository] (forge-visit-this-repository) ::

  This commands visits the repository at point in a separate buffer.

- Key: o [in forge-repositories-menu] (forge-list-owned-repositories) ::

  This command lists all known repositories that belong to the user in
  a separate buffer.  Here "known" means that an entry exists in the
  local database.  Only Github is supported for now.

The below options controls which repositories are considered to be
owned by the user.  They are additionally used by ~forge-fork~.

- User Option: forge-owned-accounts ::

  This is an alist of accounts that are owned by you.  This should
  include your username as well as any organization that you own.

  Each element has the form ~(ACCOUNT . PLIST)~.  The following
  properties are currently being used:

  - ~remote-name~ The default name suggested by ~forge-fork~ for a fork
    created within this account.  If unspecified, then the name of the
    account is used.

  Example: ~(("tarsius") ("emacsmirror" remote-name "mirror"))~.

- User Option: forge-owned-ignored ::

  This is a list of repository names that are considered to not be
  owned by you, even though they would have been considered to be
  owned by you based on ~forge-owned-accounts~.

** Exiting menus and lists
:PROPERTIES:
:UNNUMBERED: notoc
:END:

To exit a menu, type ~C-g~.  If the menu was invoked from another menu
and that menu is useful in the current buffer, then that menu becomes
active again.  If that happens and you actually want to quit all
menus, then just type ~C-g~ again.  You can also directly exit all menus
by using ~C-q~, instead of ~C-g~.

Type ~q~ to quit not only the menu, but also the list or topic detail
buffer.  That binding is also available when no menu is active, in
which case it will simply quit the buffer.  When invoked from a menu,
then this binding may return to another list buffer, in which case
some menu may also remain active.

** Default topic filters
:PROPERTIES:
:UNNUMBERED: notoc
:END:

- User Option: forge-list-buffer-default-topic-filters ::

  This option specifies the filters initially used to limit topics
  listed in topic list buffers.

- User Option: forge-status-buffer-default-topic-filters ::

  This option specifies the filters initially used to limit topics
  listed in Magit status buffers.

  Also see [[*Topic sections in Magit status buffers]].

** Topic sections in Magit status buffers
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Forge arranges for certain issues and pull-requests to be list in
Magit status buffers, by adding the following functions to
~magit-status-sections-hook~.

Which topics are listed initially is customizable using option
~forge-status-buffer-default-topic-filters~ and can be changed
temporarily for the current buffer, using ~N m f~ (~forge-topics-menu~).

- Function: forge-insert-discussions ::

  This function inserts a list of discussions, by default a list of
  "active" discussions.

- Function: forge-insert-issues ::

  This function inserts a list of issues, by default a list of "active"
  issues.

- Function: forge-insert-pullreqs ::

  This function inserts a list of pull-requests, by default a list of
  "active" pull-requests.

Forge used to provide additional functions to insert hard-coded topic
subsets, but they were removed in favor of the more flexible approach
described above.  If you miss the removed sections, you can use the
new ~forge-insert-topics~ helper function to define your own section
inserter functions.  See its docstring for more information.

If you don't want any topic list sections to be displayed in Magit
status buffers, set ~forge-add-default-sections~ to ~nil~ before ~magit~ is
loaded.

* Visiting Topics

The commands, accessible from ~forge-topic-menu~ (on ~C-return~), act on
the topic at point; so this menu is useful in buffers dedicated to
listing topics and notifications (which correspond to topics), but
also in the status buffer (which also lists topics).  In buffers
dedicated to showing details about a single topic, these commands act
on that topic; so this menu can be used there too.

To switch to this menu from another menu use ~m s~.  If the cursor is on
a topic or the current buffer visits a topic.

To display details about a topic in a separate buffer and at the same
time display the topic menu, invoke ~forge-topic-menu~ with a prefix
argument, i.e., ~C-u RET~.

- Key: RET [on topic] (forge-visit-this-topic) ::

  This commands visits the topic at point in a separate buffer.
  When invoked with a prefix argument then it not only visits the
  topic in a separate buffer, it at the same time displays

- Key: N v t (forge-visit-topic) ::
- Key: N v i (forge-visit-issue) ::
- Key: N v p (forge-visit-pullreq) ::

  These commands read a topic, issue or pull-request and visit it in a
  separate buffer.

- Key: N v u (forge-visit-topic-from-url) ::

  This commands reads an URL and visits the corresponding topic.
  Normally users would yank that URL into the minibuffer, after
  having copied it from an email or a browser's address bar.

  This command is disabled in the menu by default.

- Key: C-c C-o (forge-browse) ::
- Key: o [on topic in topic list] (forge-browse-this-topic) ::
- Key: o [on repository in repository list] (forge-browse-this-repository) ::

  These commands visit the topic, issue(s), pull-request(s), post,
  branch, commit, remote, repository or blob at point in a browser.

- Command: forge-browse-commit ::
- Command: forge-browse-branch ::
- Command: forge-browse-repository ::
- Key: N b t (forge-browse-topic) ::
- Key: N b i (forge-browse-issue) ::
- Key: N b p (forge-browse-pullreq) ::
- Key: N b r (forge-browse-remote) ::
- Key: N b I (forge-browse-issues) ::
- Key: N b P (forge-browse-pullreqs) ::

  These commands read a topic, issue(s), pull-request(s), branch,
  commit, remote or repository, and open it in a browser.

- Command: forge-browse-commit ::

  This command visit a blob in a browser.

  When invoked from a blob- or file-visiting buffer, visit that blob
  without prompting.  If the region is active, try to jump to the marked
  line or lines, and highlight them in the browser.  To what extend that
  is possible depends on the forge.  When the region is not active just
  visit the blob, without trying to jump to the current line.  When
  jumping to a line, always use a commit hash as part of the URL.  From
  a file in the worktree with no active region, instead use the branch
  name as part of the URL, unless a prefix argument is used.

  When invoked from a Dired buffer, visit the blob at point without
  prompting.  If a prefix argument is used, the commit hash is
  included in the URL.

  When invoked from any other buffer, prompt the user for a branch or
  commit, and for a file.

* Creating Topics and Posts

We call both issues and pull-requests "topics".  The contributions to
the conversation are called "posts".  The initial topic description is
also called a post.

Creating a new topic or post and editing an existing post work
similarly to now creating a new commit or editing the message of an
existing commit works in Magit.  In both cases the message has to be
written in a separate buffer and then the process has to be finished
or canceled using a separate command.  The following commands drop you
into such a buffer.

- Key: N c p (forge-create-pullreq) ::
- Key: C-c C-n [on "Pull requests" section] ::

  This command creates a new pull-request for the current repository.

- Key: N c i (forge-create-issue) ::
- Key: C-c C-n [on "Issues" section] ::

  This command creates a new issue for the current repository.

- Key: C-c C-n (forge-create-post) ::
- Key: C-c C-r ::

  This command creates a new post on an existing topic.  It is only
  available in buffers that visit an existing topic.

  If the region is active and marks part of an existing post, then
  that part of the post is quoted.  When a prefix argument is used,
  then the complete post, which point is currently on, is quoted.

The following commands are available in buffers used to edit posts:

- Key: C-c C-c (forge-post-submit) ::

  This command submits the post that is being edited in the current
  buffer.

- Key: C-c C-k (forge-post-cancel) ::

  This command cancels the post that is being edited in the current
  buffer.

- Key: C-c C-e (forge-post-dispatch) ::

  This prefix command features the above two commands as suffixes,
  and when creating a pull-request also the following command.  More
  suffix commands will likely be added in the future.

- Key: C-c C-e d (forge-post-toggle-draft) ::

  This command toggles whether the pull-request being created is a
  draft.

* Editing Topics

Many details about a topic can be changed from the buffer that visits
that topic, but also from topic lists, if the cursor is placed on the
topic to be edited.  However, to edit the posts on a topic, the topic
has to be visited in its own buffer.

- Key: C-c C-e [on a post section] (forge-edit-post) ::

  This command visits an existing post in a separate buffer, it can
  only be invoked from a topic buffer, when the cursor is on the post
  to be edited.

  Editing an existing post is similar to creating a new post, as
  described in the previous section.

- Key: C-c C-k [on a post section] (forge-delete-comment) ::

  This command deletes the post the cursor is on.  The initial message
  that was written when the topic was created, cannot be deleted, only
  replies to that.

- Key: N m s (forge-topic-menu) ::
- Key: C-<return> [on a topic section] ::

  This command displays a menu used to edit details about the topic
  the cursor is on or that is being visited in the current buffer.
  E.g., it can be used to change the status of the topic or to apply
  labels to it.  Additionally it features a few commands that act on
  that topic.

Details about a topic, such as its status and labels, can
alternatively be edited by visiting the topic in its own buffer,
navigating to the header that displays the detail and then typing ~C-c
C-e~.  This older approach is still available, but it is usually much
faster to use the menu.

* Pulling

The commands that fetch forge data are available the Forge's main menu
(~forge-dispatch~ on ~N~) and from the same menu (~magit-fetch~ on ~f~) that
is used to fetch Git data.  If ~magit-pull-or-fetch~ is non-nil, then
they are also available from the ~magit-pull~ menu (on ~F~).

With Git you have to explicitly pull Git data to make it available in
the local repository.  Forge works the same; you have to explicitly
pull to pull data using the forge's API and storing in the local
database.  This is less disruptive, more reliable, familiar and easier
to understand than if Forge pulled by itself at random intervals.  It
might however mean that you occasionally invoke a command expecting
the most recent data to be available and then have to abort and pull
first.  The same can happen with Git, e.g., you might attempt to merge
a branch that you know exists but haven't actually pulled yet.

- Key: f n (forge-pull) ::
- Key: N f f ::

  This command uses a forge's API to fetch topics and other
  information about the current repository and stores the fetched
  information in the database.

  If the current repository is still untracked locally, or the current
  repository cannot be determined, this command instead behaves like
  ~forge-add-repository~, i.e., it adds the repository to the database
  and then performs the initial pull.

- Key: f N (forge-pull-notifications) ::
- Key: N f n ::

  This command uses a forge's API to fetch all notifications from that
  forge, including, but not limited to, the notifications for the
  current repository.

  Fetching notifications fetches associated topics even for repositories
  that you have not yet explicitly added to the local database.

- Key: N f t (forge-pull-topic) ::

  This command uses a forge's API to fetch a single pull-request and
  stores it in the database.  This is useful if you chose to not fetch
  all topics when you added the repository using ~forge-add-repository~.

* Branching

Forge provides commands for creating and checking out a new branch or
work tree from a pull-request.  These commands are available from the
same transient prefix commands as the suffix commands, used to create
and check out branches and work trees in a more generic fashion
(~magit-branch~ on ~b~ and ~magit-worktree~ on ~%~).

- Key: b F (forge-branch-pullreq) ::

  This command creates and configures a new branch from a pull-request,
  creating and configuring a new remote if necessary.

  The name of the local branch is the same as the name of the remote
  branch that you are being asked to merge, unless the contributor
  could not be bothered to properly name the branch before opening the
  pull-request.  The most likely such case is when you are being asked
  to merge something like "fork/master" into "origin/master".  In such
  cases the local branch will be named "pr-N", where ~N~ is the
  pull-request number.

  These variables are always set by this command:

  - ~branch.<name>.pullRequest~ is set to the pull-request number.
  - ~branch.<name>.pullRequestRemote~ is set to the remote on which the
    pull-request branch is located.
  - ~branch.<name>.pushRemote~ is set to the same remote as
    ~branch.<name>.pullRequestRemote~ if that is possible, otherwise
    it is set to the upstream remote.
  - ~branch.<name>.description~ is set to the pull-request title.
  - ~branch.<name>.rebase~ is set to ~true~ because there should be no
    merge commits among the commits in a pull-request.

  This command also configures the upstream and the push-remote of the
  local branch that it creates.

  The branch against which the pull-request was opened is always used
  as the upstream.  This makes it easy to see what commits you are
  being asked to merge in the section titled something like "Unmerged
  into origin/master".

  Like for other commands that create a branch, it depends on the
  option ~magit-branch-prefer-remote-upstream~ whether the remote branch
  itself or the respective local branch is used as the upstream, so
  this section may also be titled, e.g., "Unmerged into master".

  When necessary and possible, the remote pull-request branch is
  configured to be used as the push-target.  This makes it easy to see
  what further changes the contributor has made since you last
  reviewed their changes in the section titled something like
  "Unpulled from origin/new-feature" or "Unpulled from
  fork/new-feature".

  - If the pull-request branch is located in the upstream repository,
    then you probably have set ~remote.pushDefault~ to that repository.
    However some users like to set that variable to their personal
    fork, even if they have push access to the upstream, so
    ~branch.<name>.pushRemote~ is set anyway.

  - The push-remote is configured using ~branch.<name>.pushRemote~, even
    if the used value is identical to that of ~remote.pushDefault~, just
    in case you change the value of the latter later on.  Additionally
    the variable ~branch.<name>.pullRequestRemote~ is set to the remote
    on which the pull-request branch is located.

  - If the pull-request branch is located on the contributor's fork,
    then you, as a maintainer of the upstream repository, are usually
    allowed to push to that branch anyway.  (However, the contributor
    could explicitly disallow this, but in my experience that rarely
    happens.)

  - As mentioned above, contributors sometimes fail to use a dedicated
    branch for their pull-requests and this command is thus forced to
    make up a branch name such as "pr-313".

    Usually a maintainer would use ~magit-push-current-to-pushremote~
    (on ~p~) to push to the contributor's pull-request branch on their
    fork.  For a branch named "pr-313" that does not work. (Instead
    of pushing to "fork/main", it would unsuccessfully attempt to
    create a new branch "pr-313" on the fork.)  In such situations
    the ~magit-push~ menu offers an alternative command for pushing to
    the pull-request branch: ~forge-push-to-unnamed-pullreq~ (on ~N~).

- Key: b f (forge-checkout-pullreq) ::

  This command creates and configures a new branch from a pull-request
  the same way ~forge-branch-pullreq~ does.  Additionally it checks out
  the new branch.

- Key: Z n (forge-checkout-worktree) ::

  This command creates and configures a new branch from a pull-request
  the same way ~forge-branch-pullreq~ does.  Additionally it checks out
  the new branch, using a new working tree.

- User Option: forge-checkout-worktree-read-directory-function ::

  This function is used by ~forge-checkout-worktree~, to read the new
  worktree directory where it checks out the pull-request.  It takes
  the pull-request as the only argument and must return a directory.

When you delete a pull-request branch, which was created using one of
the above three commands, then ~magit-branch-delete~ usually offers to
also delete the corresponding remote.  It does not offer to delete a
remote if (1) the remote is the upstream remote, and/or (2) if other
branches are being fetched from the remote.

Note that you have to delete the local branch (e.g., "feature") for
this to work.  If you delete the tracking branch (e.g., "fork/feature"),
then the remote is never removed.

* Miscellaneous Commands

- Key: N M (forge-merge) ::
- Key: m M [if enabled] ::

  This command merges the current pull-request using the forge's API.
  If there is no current pull-request or with a prefix argument, then
  it reads a pull-request to visit instead.

  The "merge method" to be used is read from the user.

  Use of this command is discouraged.  Unless the remote repository is
  configured to disallow that, you should instead merge locally and
  then push the target branch.  Forges detect that you have done that
  and respond by automatically marking the pull-request as merged.

- Key: N c f (forge-fork) ::

  This command adds an additional remote to the current repository.
  The remote can either point at an existing repository or one that
  has to be created first by forking it to an account the user has
  access to.

  Currently this only supports Github and Gitlab.

- Key: N - H (forge-toggle-topic-legend) ::

  This command toggle whether to show a legend for faces used in topic
  menus and lists.

- Key: N - S (forge-toggle-display-in-status-buffer) ::

  This command toggles whether any topics are displayed in the current
  Magit status buffer.

- Key: C-c C-w (forge-copy-url-at-point-as-kill) ::

  This command copies the url for the topic, issue(s),
  pull-request(s), post, branch, commit, remote or repository to the
  kill-ring.

  This determines the url the same way as ~forge-browse~ does, but then
  adds it to the kill-ring, instead of visiting it in a browser.

- Key: M b r (forge-rename-default-branch) ::

  This command rename the default branch to a new name read from the
  user.

  This changes the name on the upstream remotely and locally, and
  update the upstream remotes of local branches accordingly.

- Command: forge-add-pullreq-refspec ::

  This command configures Git to fetch all pull-requests.

  This is done by adding ~+refs/pull/*/head:refs/pullreqs/*~ to the
  value of ~remote.REMOTE.fetch~, where REMOTE is the upstream remote.

- Command: forge-add-user-repositories ::

  This command reads a host and a username from the user and adds all
  of that user's repositories on that host to the local database.

  This may take a while. Only Github is supported at the moment.

- Command: forge-add-organization-repositories ::

  This command reads a host and an organization from the user and adds
  all the organization's repositories on that host to the local database.

  This may take a while. Only Github is supported at the moment.

- Command: forge-remove-repository ::

  This command reads a repository and removes it from the local
  database.

- Command: forge-remove-topic-locally ::

  This command reads a topic and removes it from the local database.
  When the region marks multiple topics, then offer to remove them all.

  The topic is not removed from the forge and, if it is later modified,
  then it will be added to the database again when fetching all topics.

  This is useful for users who only fetch individual topics and want to
  remove the topics they are no longer interested in.  This can also be
  used to remove topics locally, which have already been removed on the
  forge (the service).  Forge (the package) cannot automatically detect
  when that happens, because given how the APIs work, this would be too
  expensive.

- Command: forge-reset-database ::

  This command moves the current database file to the trash and
  creates a new empty database.

  This is useful after the database's table schemata have changed,
  which will happen a few times while the Forge functionality is still
  under heavy development.

* Miscellaneous Options

- User Option: forge-database-file ::

  This option specifies the file used to store the forge database.

- User Option: forge-topic-wash-title-hook ::

  Functions used to highlight parts of each individual topic title.

  These functions are called in order, in a buffer that containing the
  topic title.  They should set text properties as they see fit, usually
  just ~font-lock-face~.  Before each function is called, point is at the
  beginning of the buffer.

- User Option: forge-topic-repository-slug-width ::

  This option specifies the width of repository slugs (i.e.,
  "OWNER/NAME").

- User Option: forge-buffer-draft-p ::

  This option controls whether new pull-requests start out as drafts
  by default.

  The buffer-local value of this variable is used to keep track of the
  draft status of the current pull-request.

- User Option: forge-repository-list-columns ::

  This option specifies the list of columns displayed when listing
  repositories.

  Each element has the form ~(HEADER SOURCE WIDTH SORT PROPS)~.

  HEADER is the string displayed in the header.  WIDTH is the width
  of the column.  SOURCE is used to get the value, it has to be the
  name of a slot of ~forge-repository~ or a function that takes
  such an object as argument.  SORT is a boolean or a function used
  to sort by this column.  Supported PROPS include ~:right-align~
  and ~:pad-right~.

- User Option: forge-limit-topic-choices ::

  This option controls whether to initially limit completion
  candidates to active topics.

- User Option: forge-post-heading-format ::

  This option specifies the format for post headings in topic view.

  The following ~%~-sequences are supported:

  - ~%a~ The forge nickname of the author.
  - ~%c~ The absolute creation date.
  - ~%C~ The relative creation date.

- User Option: forge-post-fill-region ::

  This option controls whether to call ~fill-region~ before displaying
  forge posts.

- User Option: forge-bug-reference-hooks ::

  This option lists the hooks to which ~forge-bug-reference-setup~ is
  added.  It has to be customized before ~forge~ is loaded, or it won't
  take effect.

* How Forge Detection Works
:PROPERTIES:
:APPENDIX:   t
:END:

Forge uses the Ghub package to communicate with forge APIs.  For more
information about Ghub, see [[info:ghub]].

Ghub does *not* associate a given local repository with a repository on
a forge.  The Forge package itself takes care of this.  In doing so it
ignores the Git variable ~ghub.host~ and other ~*.host~ variables used by
Ghub.  (But ~github.user~, and other variables used to specify the user,
are honored).

Forge associates the local repository with a forge repository, by first
determining which remote is associated with the upstream repository,
and then looking that up in ~forge-alist~.

If only one remote exists, then Forge uses that unconditionally.  To
reduce the number of support requests, this is even the case if the
Git variable ~forge.remote~ names another, non-existent, remote.

If several remotes exist, then a remote may be selected based on its
name.  Almost always we want to fetch the data associated with the
upstream repository, so that is what the logic described here tries to
achieve.  The convention is to name the upstream remote "origin", and
if that convention were universally followed, then things would be
trivial.  However many people name the upstream remote "upstream",
which also makes sense.

#+begin_quote
Note, however, that even though a surprising number of people do just
that, it does not make any sense to use the name "origin" to refer to
a fork; not even to your own fork.  A fork is a *copy* of the original,
"copy" is an antonym for "original", and the word "origin" is not only
closely related to but is even contained in the word "original".
Naming a fork the "origin" is at best extremely confusing.

- copy :: a thing made to be similar or identical to another.

- original :: the earliest form of something, from which copies may be
  made.

- origin :: the point or place where something begins, arises, or is
  derived.
#+end_quote

If several remotes exist, then the following remote names are tried in
order and the first remote thus named that exists in the repository is
used.

1. The value of the Git variable ~forge.remote~, if set.  If the
   variable has a value but no remote by the specified name exists,
   then a warning is shown, but otherwise this conflict is ignored.
   This behavior is arguably odd, but due to historic and pragmatic
   reasons it is the least painful path forward.

2. The remote named ~upstream~, if it exists.

3. The remote named ~origin~, if it exists.

The remote named "upstream" is preferred over the remote named
"origin" because the existence of the former strongly suggests that
the latter is either not used in this repository (in which case the
order does not matter) or else it is abused as the name of a fork (in
which case "upstream" must be preferred).

- Variable: forge.remote ::

  The value of this variable specifies the remote from which Forge
  fetches data.  It is usually best to leave this unspecified and to
  rely on the behavior described above.

  If the remote has to be specified explicitly, then this should be
  done locally, for a single repository.

  Only ever set this globally, if you consistently use a certain name
  to refer to the upstream repository and it isn't one of "upstream"
  or "origin", and you *never* use that name to refer to a repository
  that does *not* refer to the upstream repository.

- Key: N r (forge-forge.remote) ::

  This command changes the value of the ~forge.remote~ Git variable in
  the current repository.

If this variable is set, then Forge uses the remote by that name, if
it exists, the same way it may have used ~origin~ if the variable were
undefined.  I.e., it does not fall through to try ~origin~ if no remote
by your chosen name exists.

Once the upstream remote has been determined, Forge looks it up in
~forge-alist~, using the host part of the URL as the key.  For example,
the key for ~git@github.com:magit/forge.git~ is ~github.com~.

- User Option: forge-alist ::

  This option defines forge hosts known to Forge.

  Each entry has the form ~(GITHOST APIHOST WEBHOST CLASS)~.

  - GITHOST is the host used to access repositories on the forge using
    Git.

  - APIHOST is the host used to access the forge's API.  For some forges
    the isn't just a host, but a host followed by the path to the API's
    endpoint.

  - WEBHOST is the host used to access repositories on this forge using
    a browser.  The IDs used to identify repositories from the forge in
    the local database also derives from this value.

  - CLASS is the class to be used for repositories from the forge.

  Complications:

  - When connecting to a Github Enterprise edition whose REST API's
    end point is "<host>/v3" and whose GraphQL API's end point is
    "<host>/graphql", then use "<host>/v3" as APIHOST.  This is
    a historic accident.  See issue #174.

  - WEBHOST and CLASS cannot be changed once you have added one or
    more repositories from a forge.  Changing GITHOST and/or APIHOST
    may be possible, but should seldom be necessary.

* Supported Forges and Hosts
:PROPERTIES:
:APPENDIX:   t
:END:

Currently Forge supports two forges and three more forges partially.
Additionally it supports four semi-forges.  Support for more forges
and semi-forges can and will be added.

Both forges and semi-forges provide web interfaces for Git
repositories.  Forges additionally support pull-requests and issues
and make those and other information available using an API.

When a forge is only partially supported, then that means that only
the functionality that does not require the API is implemented, or
in other words, that the forge is only supported as a semi-forge.

A host is a particular instance of a forge.  For example the hosts
https://gitlab.com and https://salsa.debian.org are both instances of
the Gitlab forge.  Forge supports some well known hosts out of the box
and additional hosts can easily be supported by adding entries to the
option ~forge-alist~ (see [[*How Forge Detection Works]]).

For more details about the caveats mentioned below (and some others)
see also [[*Getting Started]].

** Supported Forges
*** Github
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Forge's support for Github can be considered the "reference
implementation".  Support for other forges can lag behind a bit.

**** Github Caveats

- Forge uses the Github GraphQL API when possible but has to fall
  back to use the REST API in many cases because the former is still
  rather incomplete.

- The Github GraphQL API has a hard-coded timeout on queries. The only solution
  is to reduce the number of entities we query at once, which can be done by
  adjusting either the ~forge.graphqlItemLimit~ git variable or the field "GQL
  entity limit" in a status buffer.

- Forge depends on the ~updated_at~ field being updated when
  appropriate.  For Github pull-requests at least, that is not always
  done.

**** Github Hosts

- https://github.com

*** Gitlab
:PROPERTIES:
:UNNUMBERED: notoc
:END:
**** Gitlab Caveats

- Forge cannot provide notifications because the Gitlab API does not
  expose those.

**** Gitlab Hosts

- https://gitlab.com
- https://salsa.debian.org
- https://framagit.org

** Partially Supported Forges
*** Forgejo https://forgejo.org
:PROPERTIES:
:UNNUMBERED: notoc
:END:

This is the next forge whose API will be supported.

**** Forgejo Hosts

- https://codeberg.org

*** Gitea https://gitea.io
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Once Forgejo is supported it might be fairly simple to support Gitea
too, because the former is a fork of the latter, and their APIs might
still be similar enough.

*** Gogs https://gogs.io
:PROPERTIES:
:UNNUMBERED: notoc
:END:

Once Forgejo is supported it might be fairly simple to support Gogs
too, because the Forgejo is a fork of Gitea, which is a fork of Gogs,
and their APIs might still be similar enough.

*** Bitbucket https://bitbucket.org
:PROPERTIES:
:UNNUMBERED: notoc
:END:

I don't plan to support Bitbucket's API any time soon, and it gets
less likely that I will every do it every time I look at it.

**** Bitbucket Caveats

- The API documentation is poor and initial tests indicated that the
  implementation is buggy.

- Atlassian's offering contains two very distinct implementations
  that are both called "Bitbucket".  Forge only supports the
  implementation whose only instance is available at
  https://bitbucket.org, because I only have access to that.

- Unlike all other forges, Bitbucket does not expose pull-requests
  as references in the upstream repository.  For that reason Forge
  actually treats it as a semi-forge, not as forge whose API is not
  supported yet.  This means that you cannot checkout pull-requests
  locally.  There is little hope that this will ever get fixed; the
  respective issue was opened six years ago and there has been no
  progress since: https://bitbucket.org/site/master/issues/5814.

**** Bitbucket Hosts

- https://bitbucket.org

** Supported Semi-Forges
*** Gitweb https://git-scm.com/docs/gitweb
:PROPERTIES:
:UNNUMBERED: notoc
:END:
**** Gitweb Caveats

- I could find only one public installation
  (https://git.savannah.gnu.org), which gives users the choice
  between Gitweb and Cgit.  The latter seems more popular (not
  just on this site).

*** Cgit https://git.zx2c4.com/cgit/about
:PROPERTIES:
:UNNUMBERED: notoc
:END:
**** Cgit Caveats

- Different sites use different URL schemata and some of the bigger
  sites use a fork.  For this reason Forge has to provide several
  classes to support different variations of Cgit and you have to
  look at their definitions to figure out which one is the correct
  one for a particular installation.

**** Cgit Hosts

- https://git.savannah.gnu.org/cgit
- https://git.kernel.org
- https://repo.or.cz

*** Stgit https://codemadness.org/git/stagit/file/README.html
:PROPERTIES:
:UNNUMBERED: notoc
:END:
**** Stgit Caveats

- Stgit cannot show logs for branches beside "master".  For that
  reason Forge takes users to a page listing the branches when they
  request the log for a particular branch (even for "master" whose
  log is just one click away from there).

**** Stgit Hosts

- https://git.suckless.org

*** Srht https://meta.sr.ht
:PROPERTIES:
:UNNUMBERED: notoc
:END:
**** Srht Caveats

- Srht cannot show logs for branches beside "master".  For that
  reason Forge takes users to a page listing the branches when they
  request the log for a particular branch (even for "master" whose
  log is just one click away from there).

**** Srht Hosts

- https://git.sr.ht

* FAQ
:PROPERTIES:
:APPENDIX:   t
:END:

This section lists some frequently asked questions.  Please see also
https://github.com/magit/forge/wiki/FAQ for an extended list of common
issues.

** ~error in process filter: HTTP Error: 502, "Bad gateway"~

This is a frequently occurring error.  Adding some formatting, the
full error is:

#+begin_src emacs-lisp
  error in process filter: ghub--signal-error: HTTP Error: 502,
    "Bad gateway", "/graphql",
    ((data . "null")
     (errors ((message . "Something went wrong while executing your query.
        This may be the result of a timeout, or it could be a GitHub bug.
        Please include `CC2C:4FEA:A1771C1:CBF40CE:5C33F7E5`
        when reporting this issue."))))
#+end_src

This indicates that something went wrong within Github's network.
Unfortunately the reason given is rather vague, but I believe this
usually happens when there are topics with one or two magnitudes more
posts than usual, which can cause GraphQL responses to become huge.

This can be countered in the affected repository by setting the Git
variable ~forge.graphqlItemLimit~:

#+begin_src shell
  git config --local forge.graphqlItemLimit 20
#+end_src

The default is specified using the ~ghub-graphql-items-per-request~,
which defaults to 50 (down from Github's default and maximum of 100).

Fetching less items per request results in more requests, which slows
down the process, which is why the default should not be too small,
but for some repositories a more aggressive limit is needed.

* Keystroke Index
:PROPERTIES:
:APPENDIX:   t
:INDEX:      ky
:COOKIE_DATA: recursive
:END:
* Function and Command Index
:PROPERTIES:
:APPENDIX:   t
:INDEX:      fn
:END:
* Variable Index
:PROPERTIES:
:APPENDIX:   t
:INDEX:      vr
:END:
* Copying
:PROPERTIES:
:COPYING:    t
:END:

#+begin_quote
Copyright (C) 2018-{{{year}}} Jonas Bernoulli <emacs.forge@jonas.bernoulli.dev>

You can redistribute this document and/or modify it under the terms
of the GNU General Public License as published by the Free Software
Foundation, either version 3 of the License, or (at your option) any
later version.

This document is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.
#+end_quote

# IMPORTANT: Also update ORG_ARGS and ORG_EVAL in the Makefile.
# Local Variables:
# indent-tabs-mode: nil
# org-src-preserve-indentation: nil
# End: