|
1 It is common for machines (as opposed to humans) to consume Mercurial. |
|
2 This help topic describes some of the considerations for interfacing |
|
3 machines with Mercurial. |
|
4 |
|
5 Choosing an Interface |
|
6 ===================== |
|
7 |
|
8 Machines have a choice of several methods to interface with Mercurial. |
|
9 These include: |
|
10 |
|
11 - Executing the ``hg`` process |
|
12 - Querying a HTTP server |
|
13 - Calling out to a command server |
|
14 |
|
15 Executing ``hg`` processes is very similar to how humans interact with |
|
16 Mercurial in the shell. It should already be familar to you. |
|
17 |
|
18 :hg:`serve` can be used to start a server. By default, this will start |
|
19 a "hgweb" HTTP server. This HTTP server has support for machine-readable |
|
20 output, such as JSON. For more, see :hg:`help hgweb`. |
|
21 |
|
22 :hg:`serve` can also start a "command server." Clients can connect |
|
23 to this server and issue Mercurial commands over a special protocol. |
|
24 For more details on the command server, including links to client |
|
25 libraries, see https://mercurial.selenic.com/wiki/CommandServer. |
|
26 |
|
27 :hg:`serve` based interfaces (the hgweb and command servers) have the |
|
28 advantage over simple ``hg`` process invocations in that they are |
|
29 likely more efficient. This is because there is significant overhead |
|
30 to spawn new Python processes. |
|
31 |
|
32 .. tip:: |
|
33 |
|
34 If you need to invoke several ``hg`` processes in short order and/or |
|
35 performance is important to you, use of a server-based interface |
|
36 is highly recommended. |
|
37 |
|
38 Environment Variables |
|
39 ===================== |
|
40 |
|
41 As documented in :hg:`help environment`, various environment variables |
|
42 influence the operation of Mercurial. The following are particularly |
|
43 relevant for machines consuming Mercurial: |
|
44 |
|
45 HGPLAIN |
|
46 If not set, Mercurial's output could be influenced by configuration |
|
47 settings that impact its encoding, verbose mode, localization, etc. |
|
48 |
|
49 It is highly recommended for machines to set this variable when |
|
50 invoking ``hg`` processes. |
|
51 |
|
52 HGENCODING |
|
53 If not set, the locale used by Mercurial will be detected from the |
|
54 environment. If the determined locale does not support display of |
|
55 certain characters, Mercurial may render these character sequences |
|
56 incorrectly (often by using "?" as a placeholder for invalid |
|
57 characters in the current locale). |
|
58 |
|
59 Explcitly setting this environment variable is a good practice to |
|
60 guarantee consistent results. "utf-8" is a good choice on UNIX-like |
|
61 environments. |
|
62 |
|
63 HGRCPATH |
|
64 If not set, Mercurial will inherit config options from config files |
|
65 using the process described in :hg:`help config`. This includes |
|
66 inheriting user or system-wide config files. |
|
67 |
|
68 When utmost control over the Mercurial configuration is desired, the |
|
69 value of ``HGRCPATH`` can be set to an explicit file with known good |
|
70 configs. In rare cases, the value can be set to an empty file or the |
|
71 null device (often ``/dev/null``) to bypass loading of any user or |
|
72 system config files. Note that these approaches can have unintended |
|
73 consequences, as the user and system config files often define things |
|
74 like the username and extensions that may be required to interface |
|
75 with a repository. |
|
76 |
|
77 Consuming Command Output |
|
78 ======================== |
|
79 |
|
80 It is common for machines to need to parse the output of Mercurial |
|
81 commands for relevant data. This section describes the various |
|
82 techniques for doing so. |
|
83 |
|
84 Parsing Raw Command Output |
|
85 -------------------------- |
|
86 |
|
87 Likely the simplest and most effective solution for consuming command |
|
88 output is to simply invoke ``hg`` commands as you would as a user and |
|
89 parse their output. |
|
90 |
|
91 The output of many commands can easily be parsed with tools like |
|
92 ``grep``, ``sed``, and ``awk``. |
|
93 |
|
94 A potential downside with parsing command output is that the output |
|
95 of commands can change when Mercurial is upgraded. While Mercurial |
|
96 does generally strive for strong backwards compatibility, command |
|
97 output does occasionally change. Having tests for your automated |
|
98 interactions with ``hg`` commands is generally recommended, but is |
|
99 even more important when raw command output parsing is involved. |
|
100 |
|
101 Using Templates to Control Output |
|
102 --------------------------------- |
|
103 |
|
104 Many ``hg`` commands support templatized output via the |
|
105 ``-T/--template`` argument. For more, see :hg:`help templates`. |
|
106 |
|
107 Templates are useful for explicitly controlling output so that |
|
108 you get exactly the data you want formatted how you want it. For |
|
109 example, ``log -T {node}\n`` can be used to print a newline |
|
110 delimited list of changeset nodes instead of a human-tailored |
|
111 output containing authors, dates, descriptions, etc. |
|
112 |
|
113 .. tip:: |
|
114 |
|
115 If parsing raw command output is too complicated, consider |
|
116 using templates to make your life easier. |
|
117 |
|
118 The ``-T/--template`` argument allows specifying pre-defined styles. |
|
119 Mercurial ships with the machine-readable styles ``json`` and ``xml``, |
|
120 which provide JSON and XML output, respectively. These are useful for |
|
121 producing output that is machine readable as-is. |
|
122 |
|
123 .. important:: |
|
124 |
|
125 The ``json`` and ``xml`` styles are considered experimental. While |
|
126 they may be attractive to use for easily obtaining machine-readable |
|
127 output, their behavior may change in subsequent versions. |
|
128 |
|
129 These styles may also exhibit unexpected results when dealing with |
|
130 certain encodings. Mercurial treats things like filenames as a |
|
131 series of bytes and normalizing certain byte sequences to JSON |
|
132 or XML with certain encoding settings can lead to surprises. |
|
133 |
|
134 Command Server Output |
|
135 --------------------- |
|
136 |
|
137 If using the command server to interact with Mercurial, you are likely |
|
138 using an existing library/API that abstracts implementation details of |
|
139 the command server. If so, this interface layer may perform parsing for |
|
140 you, saving you the work of implementing it yourself. |
|
141 |
|
142 Output Verbosity |
|
143 ---------------- |
|
144 |
|
145 Commands often have varying output verbosity, even when machine |
|
146 readable styles are being used (e.g. ``-T json``). Adding |
|
147 ``-v/--verbose`` and ``--debug`` to the command's arguments can |
|
148 increase the amount of data exposed by Mercurial. |
|
149 |
|
150 An alternate way to get the data you need is by explicitly specifying |
|
151 a template. |
|
152 |
|
153 Other Topics |
|
154 ============ |
|
155 |
|
156 revsets |
|
157 Revisions sets is a functional query language for selecting a set |
|
158 of revisions. Think of it as SQL for Mercurial repositories. Revsets |
|
159 are useful for querying repositories for specific data. |
|
160 |
|
161 See :hg:`help revsets` for more. |
|
162 |
|
163 share extension |
|
164 The ``share`` extension provides functionality for sharing |
|
165 repository data across several working copies. It can even |
|
166 automatically "pool" storage for logically related repositories when |
|
167 cloning. |
|
168 |
|
169 Configuring the ``share`` extension can lead to significant resource |
|
170 utilization reduction, particularly around disk space and the |
|
171 network. This is especially true for continuous integration (CI) |
|
172 environments. |
|
173 |
|
174 See :hg:`help -e share` for more. |