ExternalVendorCode/serval-dna
1
0
Fork 0
mirror of https://github.com/servalproject/serval-dna.git synced 2024-12-30 09:58:55 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
f78a4d69a2
serval-dna/conf_schema.h

503 lines
24 KiB
C
Raw Normal View History

Improve new config prototype code Replace ARRAY() macro with ARRAY_ATOM(), ARRAY_STRING(), ARRAY_NODE() and ARRAY_STRUCT(). Change parse function return value to flag bits, add CFEMPTY, rename CFMISSING to CFINCOMPLETE, improve logic around mandatory struct items. Renamed struct config_node to struct cf_om_node.
2012-11-26 05:42:46 +00:00
/*
Serval DNA configuration
Copyright (C) 2012 Serval Project Inc.
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.
This program 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.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
/* This file contains definitions for the schema of the Serval DNA configuration file. See comments
* in "config.h" for a description of the internal configuration API.
*
* A configuration schema is set of nested structures and arrays. By convention, the top level, or
* root of the schema is called "main", but every structure and array has its own complete API which
* can be used by itself. So if there were two independent configuration files, both could be
* defined in this file, each with a conventional name for its own root element.
*
Improve new config prototype code Add label parser functions to ARRAY schema, use to enforce numeric argv labels numbered from 1.
2012-11-27 07:40:51 +00:00
* A configuration file consists of lines of the form:
*
* FULLKEY "=" VALUE "\n"
*
* where FULLKEY has the form KEY [ "." KEY [ "." KEY [ ... ] ] ] and VALUE is any string not
* containing newline. Lines ending with "\r\n" have the "\r" stripped from the end of VALUE.
* Otherwise VALUE is preserved exactly, with all leading and trailing spaces intact.
*
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* To describe a configuration file that looks like this:
*
* some.thing.element1=integer
* some.thing.element2=string
* some.list.foo.element1=integer
* some.list.foo.element2=string
* some.list.bar.element1=integer
* some.list.bar.element2=string
* another_thing=http://my.host.com:1234/path/to/nowhere
*
* the following schema would do:
*
* STRUCT(happy)
Fix some config schema declarations Replace 'int' representation with 'uint32_nonneg', so that default value of -1 is out of reach of parser.
2013-02-27 06:00:23 +00:00
* ATOM(int32_t, element1, 0, int32_nonneg,, "An integer >= 0")
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* STRING(80, element2, "boo!", str_nonempty, MANDATORY, "A non-empty string")
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* END_STRUCT
*
Improve new config code, SORTED and NO_DUPLICATE flags Config Object Model parser now returns bitmask result of CFxxx flags and only allocates root node if the config file is non-empty. Added emalloc_zero().
2012-11-30 02:02:30 +00:00
* ARRAY(joy,)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* KEY_STRING(3, happy, str)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* VALUE_SUB_STRUCT(happy)
* END_ARRAY(16)
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
*
* STRUCT(love)
* SUB_STRUCT(happy, thing,)
* SUB_STRUCT(joy, list,)
* END_STRUCT
*
* STRUCT(main)
* SUB_STRUCT(love, some,)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* STRING(128, another_thing, "", uri,, "URL; protocol://hostname[:port][/path]")
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* END_STRUCT
*
* which would produce an API based on the following definitions (see "config.h" for more
* information):
*
* struct config_happy {
* int32_t element1;
* char element2[81];
* };
* struct config_joy {
* unsigned ac;
Improve new config prototype code The dna.helper.argv validation function now sorts the array in numerical order of labels using qsort(3). Add LABELLEN argument to ARRAY schema macros, to control the size of the label[] string (was fixed at 41 chars before).
2012-11-28 19:17:52 +00:00
* struct config_joy__element {
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* char key[4];
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* struct config_happy value;
* } av[16];
* };
* struct config_love {
* struct config_happy thing;
* struct config_joy list;
* };
* struct config_main {
* struct config_love some;
* char another_thing[129];
* };
*
* A schema definition is composed from the following STRUCT and ARRAY definitions:
*
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
* STRUCT(name [, validatorfunc])
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* element-declaration
* element-declaration
* ...
* END_STRUCT
*
* where each element-declaration is one of:
*
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* ATOM(type, element, default, repr, flags, comment)
* NODE(type, element, default, repr, flags, comment)
* STRING(strlen, element, default, repr, flags, comment)
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
* SUB_STRUCT(structname, element, flags [, default_label])
* NODE_STRUCT(structname, element, repr, flags [, default_label])
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
*
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
* ARRAY(name, flags [, validatorfunc])
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* key-declaration
* value-declaration
* END_ARRAY(size)
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
*
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* where key-declaration is one of:
*
Improve config schema declaration Remove array key SORTED flag; all arrays are now sorted by key Use representation's comparison function for sorting arrays, not a schema-supplied comparison function.
2013-03-01 04:57:28 +00:00
* KEY_ATOM(type, repr)
* KEY_STRING(strlen, repr)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
*
* and value-declaration is one of:
*
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* VALUE_ATOM(type, repr)
* VALUE_STRING(strlen, repr)
* VALUE_NODE(type, repr)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* VALUE_SUB_STRUCT(structname)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* VALUE_NODE_STRUCT(structname, repr)
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
*
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
* For defining alternative STRUCT default settings:
*
* STRUCT_DEFAULT(name, default_label)
* default-declaration
* default-declaration
* ...
* END_STRUCT_DEFAULT
*
* where each default-declaration is one of:
*
* ATOM_DEFAULT(element, default)
* STRING_DEFAULT(element, default)
* SUB_STRUCT_DEFAULT(structname, element [, default_label])
*
* Every structure defined by STRUCT has its own, native default values as given by
* the 'default' parameter in each of its element declarations. The STRUCT_DEFAULT
* declaration defines a variation on the default values that can be used to give the
* structure alternative defaults when included as a SUB_STRUCT (or NODE_STRUCT) from
* within another structure. Not all the STRUCT's elements have to be defined within
* a STRUCT_DEFAULT definition; any omitted ones take the STRUCT's native default.
*
* For defining that one STRUCT can be treated as a subset of another:
*
* STRUCT_ASSIGN(substruct, superstruct)
* element-declaration
* element-declaration
* ...
* END_STRUCT_ASSIGN
*
* where element-declaration is exactly as for STRUCT(...) above.
*
* This generates a structure copy function that copies the given elements of 'substruct'
* into the equivalent elements of 'superstruct'; ie, both structures must have the same
* element names and types.
*
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* The meanings of the parameters are:
*
* 'name'
* A label used to qualify this struct/array's API from the API components of other structs and
* arrays. This label does not appear anywhere in the config file itself; it is purely for
* internal code-related purposes.
Improve new config prototype code The dna.helper.argv validation function now sorts the array in numerical order of labels using qsort(3). Add LABELLEN argument to ARRAY schema macros, to control the size of the label[] string (was fixed at 41 chars before).
2012-11-28 19:17:52 +00:00
* 'strlen'
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* For STRING, LABEL_STRING and VALUE_STRING, gives the maximum length of the string. The
* string is declared as an array of char[strlen+1] to leave room for a terminating nul.
Improve new config prototype code Add optional validator functions for STRUCT and ARRAY parsing. Tighten up logic for parsing STRUCTs and ARRAYs and the result flags. Improve documentation in block comments. Add CFUNSUPPORTED result flag.
2012-11-27 06:17:30 +00:00
* 'size'
* For all ARRAYs, gives the maximum size of the array.
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* 'type'
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* Used for ATOM, NODE, LABEL_ATOM, VALUE_ATOM and VALUE_NODE declarations. Gives the C type
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* of the element. For STRING, KEY_STRING and VALUE_STRING this is implicitly char[strlen+1].
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* 'structname'
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
* Only used for SUB_STRUCT, NODE_STRUCT, VALUE_SUB_STRUCT and VALUE_NODE_STRUCT declarations.
* Identifies a sub- structure by 'name' to nest in the enclosing struct or array.
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* 'element'
* The name of the struct element and the key in the configuration file. This name does appear
* in the config file and also in the API, so that an option mamed "some.thing.element1" in the
* file is referred to as some.thing.element1 in the C code. Arrays are more complicated:
* "some.list.foo.element1" in the config file is referred to as some.list.av[n].value.element1
* in the C code, and some.list.ac gives the size of the some.list.av array.
* 'default'
* Only used for ATOM and NODE struct elements. Gives the default value for the element if
* absent from the config file.
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* 'repr'
Add config cf_cmp_ functions
2013-02-27 07:41:18 +00:00
* The string representation. This name specifies a trio of functions, cf_opt_<repr>(),
* cf_fmt_<repr>() and cf_cmp_<repr>():
* - The cf_opt_<repr>() functions for ATOM, STRING, KEY_ATOM, KEY_STRING, VALUE_ATOM and
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
* VALUE_STRING take a (const char *) argument pointing to nul-terminated text. The
* <repr> functions for NODE and VALUE_NODE take a pointer to a COM node (const
* struct cf_om_node *), and are responsible for parsing the node's text and all of its
* descendents (children).
Add config cf_cmp_ functions
2013-02-27 07:41:18 +00:00
* - Each cf_fmt_<repr>() function is the inverse of cf_opt_<repr>. The cf_fmt_<repr>
* functions for ATOM, KEY_ATOM, KEY_STRING, VALUE_ATOM and VALUE_STRING all take a pointer
* to a const 'type', and produce a malloc()ed nul-terminated string which, if passed to
* cf_opt_<repr>(), would produce the same original value. If the value is invalid (outside
* the legal range) then cf_fmt_<repr> returns CFINVALID or CFEMPTY. The cf_fmt_<repr>
* functions for NODE and VALUE_NODE take a pointer to a const 'type' and produce a
* malloc()ed COM node (struct cf_om_node *) which, if passed to cf_opt_<repr> would produce
* the same original value.
* - Each cf_cmp_<repr>() function compares two values of the given 'type' and returns -1, 0
* or 1 to indicate the natural ordering of the values. These functions are used to detect
* when config elements have their default values, to avoid calling cf_fmt_<repr>(). They
* are also used to sort array keys.
Improve new config prototype code Add optional validator functions for STRUCT and ARRAY parsing. Tighten up logic for parsing STRUCTs and ARRAYs and the result flags. Improve documentation in block comments. Add CFUNSUPPORTED result flag.
2012-11-27 06:17:30 +00:00
* 'validatorfunc'
* A function that is called after the struct/array is fully parsed and populated. This
* function can perform validation checks on the whole struct/array that cannot be performed by
* the parse functions of each element in isolation, and can even alter the contents of the
* struct/array, eg, sort an array or fill in default values in structs that depend on other
* elements. Takes as its second argument the CFxxx code produced by the parser, and returns
* an updated CFxxx result code (which could be the same) as documented in "config.h".
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
* 'flags'
* A space-separated list of flags. At present only the MANDATORY flag is supported, which
* will cause parsing to fail if the given STRUCT element is not set in the config file. In
* the case of struct elements that are arrays, the config file must set at least one element
* of the array, or parsing fails.
* 'comment'
* A human-readable string describing the value of the configuration option. Must be
* informative enough to help users diagnose parse errors. Eg, "An integer" is not enough;
* better would be "Integer >= 0, number of seconds since Unix epoch".
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
* 'default_label'
* A label used to qualify an alternative STRUCT default set. These labels need only be unique
* for each given struct 'name', so different STRUCTs may re-use the same labels.
Add documentation comments for new config code
2012-11-26 05:45:59 +00:00
*
* @author Andrew Bettison <andrew@servalproject.com>
*/
Overhaul debug flags Replace debugflags_t and DEBUG_XXX bit masks with config schema "debug.xxx" entries. No more support for "debug.all".
2012-12-11 05:29:46 +00:00
STRUCT(debug)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, verbose, 0, boolean,, "")
Split debugging info related to ACK's and packet retransmissions
2013-08-30 05:26:22 +00:00
ATOM(bool_t, ack, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, dnaresponses, 0, boolean,, "")
ATOM(bool_t, dnahelper, 0, boolean,, "")
ATOM(bool_t, queues, 0, boolean,, "")
ATOM(bool_t, timing, 0, boolean,, "")
Refactor to rename "Rhizome HTTP server" to "HTTPD server" Move HTTPD start/stop/dispatch logic from rhizome_httpd.c into new httpd.c Rename config.debug.httpd -> config.debug.http_server Rename config.debug.rhizome_httpd -> config.debug.httpd Rename config.debug.rhizome_nohttptx -> config.debug.nohttptx
2014-01-31 07:32:41 +00:00
ATOM(bool_t, http_server, 0, boolean,, "")
Fix basic bugs in new HTTP server code All 'rhizomeprotocol' HTTP tests still fail
2013-10-25 07:08:51 +00:00
ATOM(bool_t, httpd, 0, boolean,, "")
Refactor to rename "Rhizome HTTP server" to "HTTPD server" Move HTTPD start/stop/dispatch logic from rhizome_httpd.c into new httpd.c Rename config.debug.httpd -> config.debug.http_server Rename config.debug.rhizome_httpd -> config.debug.httpd Rename config.debug.rhizome_nohttptx -> config.debug.nohttptx
2014-01-31 07:32:41 +00:00
ATOM(bool_t, nohttptx, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, io, 0, boolean,, "")
ATOM(bool_t, verbose_io, 0, boolean,, "")
Add debug output for mavlink sent and received frames. debug.interactive_io can be used to pause after sending or receiving a packet.
2013-09-01 08:03:32 +00:00
ATOM(bool_t, interactive_io, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, packetformats, 0, boolean,, "")
ATOM(bool_t, gateway, 0, boolean,, "")
ATOM(bool_t, keyring, 0, boolean,, "")
ATOM(bool_t, security, 0, boolean,, "")
ATOM(bool_t, mdprequests, 0, boolean,, "")
Add msp debug flag
2013-12-20 02:02:47 +00:00
ATOM(bool_t, msp, 0, boolean,, "")
Introduce basic vomp console test
2014-02-03 05:21:48 +00:00
ATOM(bool_t, monitor, 0, boolean,, "")
Rename mavlink to radio_link - split link state from slip structures and encapsulate it completely - allocate interface specific buffers on demand
2013-10-30 06:39:37 +00:00
ATOM(bool_t, radio_link, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, peers, 0, boolean,, "")
Improve overlay buffer ob_xxx() primitives All ob_append_xxx(b,...) functions return void ob_makespace() returns 1 if successful, 0 if not Add ob_overrun(b) predicate to check for overrun after any number of appends
2013-11-25 06:13:32 +00:00
ATOM(bool_t, overlaybuffer, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, overlayframes, 0, boolean,, "")
ATOM(bool_t, overlayabbreviations, 0, boolean,, "")
ATOM(bool_t, overlayrouting, 0, boolean,, "")
ATOM(bool_t, overlayroutemonitor, 0, boolean,, "")
ATOM(bool_t, overlayinterfaces, 0, boolean,, "")
ATOM(bool_t, broadcasts, 0, boolean,, "")
ATOM(bool_t, packettx, 0, boolean,, "")
ATOM(bool_t, packetrx, 0, boolean,, "")
ATOM(bool_t, packetradio, 0, boolean,, "")
ATOM(bool_t, rejecteddata, 0, boolean,, "")
ATOM(bool_t, slip, 0, boolean,, "")
ATOM(bool_t, slipdecode, 0, boolean,, "")
add new debug option to show packet radio byte stream. make uartbps default to 57600.
2013-05-07 07:59:11 +00:00
ATOM(bool_t, slipbytestream, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, packetconstruction, 0, boolean,, "")
ATOM(bool_t, rhizome, 0, boolean,, "")
Refactor manifest: specific setter functions Replace generic rhizome_manifest_set() and rhizome_manifest_set_ll() with per-field setter functions, eg, rhizome_manifest_set_filesize(). Struct rhizome_manifest elements for all known fields, to replace the use of rhizome_manifest_get() and rhizome_manifest_get_ll() everywhere: sender, recipient, service, name, date, bundle_key. Add boolean validity flags for binary blob types, to avoid having to compare with many bytes of all-zero to detect presence, eg, has_sender, has_recipient, has_author, has_bundle_key. These maintained by the setter functions. Rename existing manifest struct elements to be the same as their field names: fileLength -> filesize, journalTail -> tail. More use of unsigned int, size_t and uint64_t for payload sizes, offsets, byte counts, etc. especially in rhizome_store.c and meshms.c. More uniform use of size_t to dimension memory buffers. Fix some printf(3) style format strings for 64-bit correctness on 32-bit systems. Use new constant RHIZOME_SIZE_UNSET instead of -1 to indicate unknown dimension, and explicitly assert its absence before comparisons and arithmetic, for safety. Replace some 'int' loop variables with 'unsigned' where appropriate. Fix bugs discovered in MeshMS bundle private/public key generation and bundle secret key handling for export/extract commands. Instrument the first MeshMS test case to aid debugging. New debug config flag: debug.manifest logs all modifications to all manifest fields by setter functions. Rename debug config flag: debug.rhizome_bind -> debug.rhizome_sql_bind.
2013-10-30 12:52:19 +00:00
ATOM(bool_t, rhizome_manifest, 0, boolean,, "")
ATOM(bool_t, rhizome_sql_bind, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, rhizome_tx, 0, boolean,, "")
ATOM(bool_t, rhizome_rx, 0, boolean,, "")
ATOM(bool_t, rhizome_ads, 0, boolean,, "")
add extra debug option to trace MDP Rhizome transfer problems.
2013-08-25 11:46:33 +00:00
ATOM(bool_t, rhizome_mdp_rx, 0, boolean,, "")
Add 'debug.subscriber' config option Add debug logging to find_subscriber() function
2013-11-25 12:10:14 +00:00
ATOM(bool_t, subscriber, 0, boolean,, "")
Add tx throttling for packet radios avoids missing data due to lack of flow control.
2013-08-27 11:13:56 +00:00
ATOM(bool_t, throttling, 0, boolean,, "")
First cut of new meshms API, unfinished
2013-07-22 08:25:02 +00:00
ATOM(bool_t, meshms, 0, boolean,, "")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, manifests, 0, boolean,, "")
ATOM(bool_t, vomp, 0, boolean,, "")
ATOM(bool_t, trace, 0, boolean,, "")
ATOM(bool_t, profiling, 0, boolean,, "")
ATOM(bool_t, externalblobs, 0, boolean,, "")
Add basic link state routing
2013-04-26 07:23:04 +00:00
ATOM(bool_t, linkstate, 0, boolean,, "")
Overhaul debug flags Replace debugflags_t and DEBUG_XXX bit masks with config schema "debug.xxx" entries. No more support for "debug.all".
2012-12-11 05:29:46 +00:00
END_STRUCT
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
#define LOG_FORMAT_OPTIONS \
ATOM(bool_t, show_pid, 1, boolean,, "If true, all log lines contain PID of logging process") \
ATOM(bool_t, show_time, 1, boolean,, "If true, all log lines contain time stamp") \
ATOM(int, level, LOG_LEVEL_DEBUG, log_level,, "Only log messages at and above this level of severity") \
Improve configuration dump to log Refactor much of the new code in log.c Add new functions: cf_load_strict() and cf_reload_strict() for use by the server, which does not want to overwrite its in-memory config at all if there is any defect in the config file. However, commands always want to overwrite the default in-memory config, even with a defective config file, so that logging has a chance to succeed. Added `log_format.dump_config` boolean config option
2013-03-28 13:52:08 +00:00
ATOM(bool_t, dump_config, 1, boolean,, "If true, current configuration is dumped into start of log")
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
STRUCT(log_format)
LOG_FORMAT_OPTIONS
Add log format config options for file and stderr Upgrade configuration schema macros to allow optional "default label" argument to SUB_STRUCT and VALUE_SUB_STRUCT elements. New STRUCT_DEFAULT section declares alternative STRUCT default values with a label.
2013-03-25 07:16:56 +00:00
END_STRUCT
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
STRUCT(log_format_file)
STRING(256, directory_path, "log", str_nonempty,, "Path of directory for log files, either absolute or relative to instance directory")
Improve logging tests, fix failures, add 'log.file.path' config option
2013-04-05 05:53:20 +00:00
STRING(256, path, "", str_nonempty,, "Path of single log file, either absolute or relative to directory_path")
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
ATOM(unsigned short, rotate, 12, ushort,, "Number of log files to rotate, zero means no deletion")
ATOM(uint32_t, duration, 3600, uint32_time_interval,, "Time duration of each log file, zero means one file per invocation")
LOG_FORMAT_OPTIONS
END_STRUCT
STRUCT_ASSIGN(log_format, log_format_file)
LOG_FORMAT_OPTIONS
END_STRUCT_ASSIGN
Add log format config options for file and stderr Upgrade configuration schema macros to allow optional "default label" argument to SUB_STRUCT and VALUE_SUB_STRUCT elements. New STRUCT_DEFAULT section declares alternative STRUCT default values with a label.
2013-03-25 07:16:56 +00:00
STRUCT(log)
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
SUB_STRUCT(log_format_file, file,,)
Improve logging: add HINT level, NOWHENCE() macro
2013-04-09 07:47:22 +00:00
SUB_STRUCT(log_format, console,, console)
Improve config schema macros and log options Add STRUCT_ASSIGN macro to config schema Document STRUCT_DEFAULT and STRUCT_ASSIGN in block comments Move 'log.file_xxx' options to 'log.file.xxx'
2013-04-05 03:37:31 +00:00
SUB_STRUCT(log_format, android,, android)
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Prototype code for new config parser
2012-11-20 07:39:19 +00:00
Improve logging: add HINT level, NOWHENCE() macro
2013-04-09 07:47:22 +00:00
STRUCT_DEFAULT(log_format, console)
Improve configuration dump to log Refactor much of the new code in log.c Add new functions: cf_load_strict() and cf_reload_strict() for use by the server, which does not want to overwrite its in-memory config at all if there is any defect in the config file. However, commands always want to overwrite the default in-memory config, even with a defective config file, so that logging has a chance to succeed. Added `log_format.dump_config` boolean config option
2013-03-28 13:52:08 +00:00
ATOM_DEFAULT(show_pid, 0)
ATOM_DEFAULT(show_time, 0)
Improve logging: add HINT level, NOWHENCE() macro
2013-04-09 07:47:22 +00:00
ATOM_DEFAULT(level, LOG_LEVEL_HINT)
Improve configuration dump to log Refactor much of the new code in log.c Add new functions: cf_load_strict() and cf_reload_strict() for use by the server, which does not want to overwrite its in-memory config at all if there is any defect in the config file. However, commands always want to overwrite the default in-memory config, even with a defective config file, so that logging has a chance to succeed. Added `log_format.dump_config` boolean config option
2013-03-28 13:52:08 +00:00
ATOM_DEFAULT(dump_config, 0)
Add log format config options for file and stderr Upgrade configuration schema macros to allow optional "default label" argument to SUB_STRUCT and VALUE_SUB_STRUCT elements. New STRUCT_DEFAULT section declares alternative STRUCT default values with a label.
2013-03-25 07:16:56 +00:00
END_STRUCT_DEFAULT
Log to file, Android and stderr simultaneously Indepent configuration of show_pid, show_time and log level for each destination Update test scripts for new config options Include xprintf.c in MDP client source files (now used by log.c)
2013-03-26 07:54:37 +00:00
STRUCT_DEFAULT(log_format, android)
Improve configuration dump to log Refactor much of the new code in log.c Add new functions: cf_load_strict() and cf_reload_strict() for use by the server, which does not want to overwrite its in-memory config at all if there is any defect in the config file. However, commands always want to overwrite the default in-memory config, even with a defective config file, so that logging has a chance to succeed. Added `log_format.dump_config` boolean config option
2013-03-28 13:52:08 +00:00
ATOM_DEFAULT(show_pid, 0)
Log to file, Android and stderr simultaneously Indepent configuration of show_pid, show_time and log level for each destination Update test scripts for new config options Include xprintf.c in MDP client source files (now used by log.c)
2013-03-26 07:54:37 +00:00
END_STRUCT_DEFAULT
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
STRUCT(server)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(256, chdir, "/", absolute_path,, "Absolute path of chdir(2) for server process")
STRING(256, interface_path, "", str_nonempty,, "Path of directory containing interface files, either absolute or relative to instance directory")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, respawn_on_crash, 0, boolean,, "If true, server will exec(2) itself on fatal signals, eg SEGV")
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
END_STRUCT
Improve new config prototype code Add monitor.socket and monitor.uid to schema. Remove some deprecated warning functions.
2012-11-28 07:43:53 +00:00
STRUCT(monitor)
Fix some config schema declarations Replace 'int' representation with 'uint32_nonneg', so that default value of -1 is out of reach of parser.
2013-02-27 06:00:23 +00:00
ATOM(uint32_t, uid, 0, uint32_nonzero,, "Allowed UID for monitor socket client")
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
END_STRUCT
STRUCT(mdp_iftype)
Fix intermittent 'routing' test failures Caused by tests running so slowly that nodes become unreachable due to timeout before the assertion of reachability is made. The timeout was hard coded at five times the tick interval (2.5 seconds). Introduced new config options: mdp.iftype.*.reachable_timeout_ms and interface.*.mdp.reachable_timeout_ms to allow the timeout to be configured. Configure the timeout to 60 seconds in 'routing' tests. Changed the config schema to move options interfaces.*.mdp_tick_ms and interfaces.*.packet_interval into interfaces.*.mdp.tick_ms and interfaces.*.mdp.packet_interval (introduced new sub-struct 'mdp').
2013-10-08 06:06:07 +00:00
ATOM(int32_t, tick_ms, -1, int32_nonneg,, "Tick interval")
Fix some config schema declarations Replace 'int' representation with 'uint32_nonneg', so that default value of -1 is out of reach of parser.
2013-02-27 06:00:23 +00:00
ATOM(int32_t, packet_interval, -1, int32_nonneg,, "Minimum interval between packets in microseconds")
Fix intermittent 'routing' test failures Caused by tests running so slowly that nodes become unreachable due to timeout before the assertion of reachability is made. The timeout was hard coded at five times the tick interval (2.5 seconds). Introduced new config options: mdp.iftype.*.reachable_timeout_ms and interface.*.mdp.reachable_timeout_ms to allow the timeout to be configured. Configure the timeout to 60 seconds in 'routing' tests. Changed the config schema to move options interfaces.*.mdp_tick_ms and interfaces.*.packet_interval into interfaces.*.mdp.tick_ms and interfaces.*.mdp.packet_interval (introduced new sub-struct 'mdp').
2013-10-08 06:06:07 +00:00
ATOM(int32_t, reachable_timeout_ms, -1, int32_nonneg,, "Inactivity timeout after which node considered unreachable")
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
END_STRUCT
Improve new config code, SORTED and NO_DUPLICATE flags Config Object Model parser now returns bitmask result of CFxxx flags and only allocates root node if the config file is non-empty. Added emalloc_zero().
2012-11-30 02:02:30 +00:00
ARRAY(mdp_iftypelist, NO_DUPLICATES)
Improve config schema declaration Remove array key SORTED flag; all arrays are now sorted by key Use representation's comparison function for sorting arrays, not a schema-supplied comparison function.
2013-03-01 04:57:28 +00:00
KEY_ATOM(short, interface_type)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
VALUE_SUB_STRUCT(mdp_iftype)
END_ARRAY(5)
STRUCT(mdp)
SUB_STRUCT(mdp_iftypelist, iftype,)
Add proc files for http server & mdp udp ports
2014-02-24 06:17:16 +00:00
ATOM(bool_t, enable_inet, 0, boolean,, "If true, allow mdp clients to connect over loopback UDP")
Improve new config prototype code Add monitor.socket and monitor.uid to schema. Remove some deprecated warning functions.
2012-11-28 07:43:53 +00:00
END_STRUCT
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
STRUCT(olsr)
Pre-emptively send small payloads via MDP
2013-08-20 23:51:40 +00:00
ATOM(bool_t, enable, 0, boolean,, "If true, OLSR is used for mesh routing")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(uint16_t, remote_port, 4130, uint16_nonzero,, "Remote port number")
ATOM(uint16_t, local_port, 4131, uint16_nonzero,, "Local port number")
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
END_STRUCT
Improve config schema declaration Remove array key SORTED flag; all arrays are now sorted by key Use representation's comparison function for sorting arrays, not a schema-supplied comparison function.
2013-03-01 04:57:28 +00:00
ARRAY(argv, NO_DUPLICATES, vld_argv)
KEY_ATOM(unsigned short, ushort_nonzero)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
VALUE_STRING(128, str)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
END_ARRAY(16)
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
Improve new config prototype code Add optional validator functions for STRUCT and ARRAY parsing. Tighten up logic for parsing STRUCTs and ARRAYs and the result flags. Improve documentation in block comments. Add CFUNSUPPORTED result flag.
2012-11-27 06:17:30 +00:00
STRUCT(executable)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(256, executable, "", absolute_path, MANDATORY, "Absolute path of dna helper executable")
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
SUB_STRUCT(argv, argv,)
END_STRUCT
STRUCT(dna)
Improve new config prototype code Add optional validator functions for STRUCT and ARRAY parsing. Tighten up logic for parsing STRUCTs and ARRAYs and the result flags. Improve documentation in block comments. Add CFUNSUPPORTED result flag.
2012-11-27 06:17:30 +00:00
SUB_STRUCT(executable, helper,)
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
END_STRUCT
Integrate new config into servald Not passing any tests yet, but compiles and links and simple uses do not SEGV.
2012-12-04 03:42:28 +00:00
STRUCT(rhizome_peer)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(25, protocol, "http", protocol,, "Protocol name")
STRING(256, host, "", str_nonempty, MANDATORY, "Host name or IP address")
Refactor to rename "Rhizome HTTP server" to "HTTPD server" Move HTTPD start/stop/dispatch logic from rhizome_httpd.c into new httpd.c Rename config.debug.httpd -> config.debug.http_server Rename config.debug.rhizome_httpd -> config.debug.httpd Rename config.debug.rhizome_nohttptx -> config.debug.nohttptx
2014-01-31 07:32:41 +00:00
ATOM(uint16_t, port, HTTPD_PORT, uint16_nonzero,, "Port number")
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Prototype code for new config parser
2012-11-20 07:39:19 +00:00
Improve new config code, SORTED and NO_DUPLICATE flags Config Object Model parser now returns bitmask result of CFxxx flags and only allocates root node if the config file is non-empty. Added emalloc_zero().
2012-11-30 02:02:30 +00:00
ARRAY(peerlist,)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
KEY_STRING(15, str)
VALUE_NODE_STRUCT(rhizome_peer, rhizome_peer)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
END_ARRAY(10)
Improve new config prototype code Make config LISTs more flexible. Start implementing existing config options: directory.service (SID parsing), interfaces (better list logic).
2012-11-22 07:50:13 +00:00
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
STRUCT(rhizome_direct)
Improve new config prototype code Refactor and rename, improve logic for reporting unsupported nodes.
2012-11-23 08:43:02 +00:00
SUB_STRUCT(peerlist, peer,)
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Improve new config prototype, add LIST to schema
2012-11-21 07:39:05 +00:00
First HTTP RESTful interface tests
2013-10-14 07:16:09 +00:00
STRUCT(user)
STRING(50, password, "", str,, "Authentication password")
END_STRUCT
ARRAY(userlist,)
KEY_STRING(25, str)
VALUE_SUB_STRUCT(user)
END_ARRAY(10)
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
STRUCT(rhizome_api_addfile)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
STRING(64, uri_path, "", absolute_path,, "URI path for HTTP add-file request")
ATOM(struct in_addr, allow_host, hton_in_addr(INADDR_LOOPBACK), in_addr,, "IP address of host allowed to make HTTP add-file request")
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(256, manifest_template_file, "", str_nonempty,, "Path of manifest template file, either absolute or relative to instance directory")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(sid_t, default_author, SID_ANY, sid,, "Author of add-file bundle if sender not given")
ATOM(rhizome_bk_t, bundle_secret_key, RHIZOME_BK_NONE, rhizome_bk,, "Secret key of add-file bundle to try if sender not given")
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
END_STRUCT
First HTTP RESTful interface tests
2013-10-14 07:16:09 +00:00
STRUCT(rhizome_api_restful)
SUB_STRUCT(userlist, users,)
Implement HTTP /restful/rhizome/newsince/.../bundlelist.json Now functions as per requirements, blocks for 60 seconds sending new bundles as they appear, sends new token with each new bundle. Tested manually.
2013-11-19 05:13:51 +00:00
ATOM(uint32_t, newsince_timeout, 60, uint32_time_interval,, "Time to block while reporting new bundles")
Add 'rhizome.api.restful.newsince_poll_ms' config option
2013-11-20 04:20:57 +00:00
ATOM(uint32_t, newsince_poll_ms, 2000, uint32_nonzero,, "Database poll interval while blocked reporting new bundles")
First HTTP RESTful interface tests
2013-10-14 07:16:09 +00:00
END_STRUCT
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
STRUCT(rhizome_api)
SUB_STRUCT(rhizome_api_addfile, addfile,)
First HTTP RESTful interface tests
2013-10-14 07:16:09 +00:00
SUB_STRUCT(rhizome_api_restful, restful,)
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
END_STRUCT
Merge branch 'newconfig' into 'master' Conflicts: commandline.c conf.h dataformats.c log.h overlay_address.c overlay_interface.c packetformats.c rhizome.c serval.h tests/directory_service vomp_console.c
2012-12-07 03:39:55 +00:00
STRUCT(rhizome_http)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, enable, 1, boolean,, "If true, Rhizome HTTP server is started")
Merge branch 'newconfig' into 'master' Conflicts: commandline.c conf.h dataformats.c log.h overlay_address.c overlay_interface.c packetformats.c rhizome.c serval.h tests/directory_service vomp_console.c
2012-12-07 03:39:55 +00:00
END_STRUCT
STRUCT(rhizome_mdp)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, enable, 1, boolean,, "If true, Rhizome MDP server is started")
Merge branch 'newconfig' into 'master' Conflicts: commandline.c conf.h dataformats.c log.h overlay_address.c overlay_interface.c packetformats.c rhizome.c serval.h tests/directory_service vomp_console.c
2012-12-07 03:39:55 +00:00
END_STRUCT
STRUCT(rhizome_advertise)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, enable, 1, boolean,, "If true, Rhizome advertisements are sent")
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
ATOM(uint32_t, interval, 500, uint32_nonzero,, "Interval between Rhizome advertisements")
Merge branch 'newconfig' into 'master' Conflicts: commandline.c conf.h dataformats.c log.h overlay_address.c overlay_interface.c packetformats.c rhizome.c serval.h tests/directory_service vomp_console.c
2012-12-07 03:39:55 +00:00
END_STRUCT
STRUCT(rhizome)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, enable, 1, boolean,, "If true, server opens Rhizome database when starting")
ATOM(bool_t, fetch, 1, boolean,, "If false, no new bundles will be fetched from peers")
Don't clean the database for every command line operation by default
2013-08-27 04:12:56 +00:00
ATOM(bool_t, clean_on_open, 0, boolean,, "If true, Rhizome database is cleaned at start of every command")
ATOM(bool_t, clean_on_start, 1, boolean,, "If true, Rhizome database is cleaned at start of daemon")
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(256, datastore_path, "", absolute_path,, "Path of rhizome storage directory, absolute or relative to instance directory")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(uint64_t, database_size, 1000000, uint64_scaled,, "Size of database in bytes")
Replace config 'rhizome.external_blobs' with 'rhizome.max_blob_size'
2013-12-30 04:21:20 +00:00
ATOM(uint32_t, max_blob_size, 128 * 1024, uint32_scaled,, "Store payloads larger than this in files not SQLite blobs")
simplify external/internal blob storage to a boolean flag instead of a threshold.
2013-02-18 04:39:03 +00:00
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
ATOM(uint64_t, rhizome_mdp_block_size, 512, uint64_scaled,, "Rhizome MDP block size.")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(uint64_t, idle_timeout, RHIZOME_IDLE_TIMEOUT, uint64_scaled,, "Rhizome transfer timeout if no data received.")
Increase default mdp timeout & make it configurable
2013-11-04 01:42:58 +00:00
ATOM(uint64_t, mdp_stall_timeout, 1000, uint64_scaled,, "Timeout to request more data via mdp.")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(uint32_t, fetch_delay_ms, 50, uint32_nonzero,, "Delay from receiving first bundle advert to initiating fetch")
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
SUB_STRUCT(rhizome_direct, direct,)
SUB_STRUCT(rhizome_api, api,)
Merge branch 'newconfig' into 'master' Conflicts: commandline.c conf.h dataformats.c log.h overlay_address.c overlay_interface.c packetformats.c rhizome.c serval.h tests/directory_service vomp_console.c
2012-12-07 03:39:55 +00:00
SUB_STRUCT(rhizome_http, http,)
SUB_STRUCT(rhizome_mdp, mdp,)
SUB_STRUCT(rhizome_advertise, advertise,)
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Improve new config prototype code Make config LISTs more flexible. Start implementing existing config options: directory.service (SID parsing), interfaces (better list logic).
2012-11-22 07:50:13 +00:00
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
STRUCT(directory)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
ATOM(sid_t, service, SID_ANY, sid,, "Subscriber ID of Serval Directory Service")
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Improve new config prototype, add LIST to schema
2012-11-21 07:39:05 +00:00
Improve new config prototype code Add hosts.SIDHEX.{interface,network,port} config options.
2012-11-28 19:58:28 +00:00
STRUCT(host)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
STRING(INTERFACE_NAME_STRLEN, interface, "", str_nonempty,, "Interface name")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
STRING(256, host, "", str_nonempty,, "Host Name")
ATOM(struct in_addr, address, hton_in_addr(INADDR_NONE), in_addr,, "Host IP address")
ATOM(uint16_t, port, PORT_DNA, uint16_nonzero,, "Port number")
Improve new config prototype code Add hosts.SIDHEX.{interface,network,port} config options.
2012-11-28 19:58:28 +00:00
END_STRUCT
Improve new config code, SORTED and NO_DUPLICATE flags Config Object Model parser now returns bitmask result of CFxxx flags and only allocates root node if the config file is non-empty. Added emalloc_zero().
2012-11-30 02:02:30 +00:00
ARRAY(host_list, NO_DUPLICATES)
Improve config schema declaration Remove array key SORTED flag; all arrays are now sorted by key Use representation's comparison function for sorting arrays, not a schema-supplied comparison function.
2013-03-01 04:57:28 +00:00
KEY_ATOM(sid_t, sid)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
VALUE_SUB_STRUCT(host)
END_ARRAY(32)
Improve new config prototype code Add hosts.SIDHEX.{interface,network,port} config options.
2012-11-28 19:58:28 +00:00
Fix new 'interfaces' config option Introduce CFINCOMPATIBLE config parse result flag. Sort interface rules by unsigned integer key. Legacy and modern 'interfaces' config styles are now incompatible. Validate config_network_interface struct to enforce that only exactly one of 'match' and 'dummy' options are set. Add test cases for 'interface' config option.
2012-12-05 02:28:07 +00:00
STRUCT(network_interface, vld_network_interface)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, exclude, 0, boolean,, "If true, do not use matching interfaces")
ATOM(struct pattern_list, match, PATTERN_LIST_EMPTY, pattern_list,, "Names that match network interface")
Improve config description
2013-09-16 05:34:08 +00:00
ATOM(short, socket_type, SOCK_UNSPECIFIED, socket_type,, "Type of network socket; stream, dgram or file")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(short, encapsulation, ENCAP_OVERLAY, encapsulation,, "Type of packet encapsulation")
STRING(256, file, "", str_nonempty,, "Path of interface file, absolute or relative to server.interface_path")
ATOM(struct in_addr, dummy_address, hton_in_addr(INADDR_LOOPBACK), in_addr,, "Dummy interface address")
ATOM(struct in_addr, dummy_netmask, hton_in_addr(0xFFFFFF00), in_addr,, "Dummy interface netmask")
ATOM(uint16_t, port, PORT_DNA, uint16_nonzero,, "Port number for network interface")
Improve reliability - don't drop packets that haven't been resolved yet on incoming acks - statistically drop unicast packets during tests
2013-08-12 04:51:31 +00:00
ATOM(bool_t, drop_broadcasts, 0, boolean,, "If true, drop all incoming broadcast packets")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, drop_unicasts, 0, boolean,, "If true, drop all incoming unicast packets")
Improve reliability - don't drop packets that haven't been resolved yet on incoming acks - statistically drop unicast packets during tests
2013-08-12 04:51:31 +00:00
ATOM(uint16_t, drop_packets, 0, uint16_nonzero,, "Percentage of incoming packets that should be dropped for testing purposes")
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(short, type, OVERLAY_INTERFACE_WIFI, interface_type,, "Type of network interface")
Fix intermittent 'routing' test failures Caused by tests running so slowly that nodes become unreachable due to timeout before the assertion of reachability is made. The timeout was hard coded at five times the tick interval (2.5 seconds). Introduced new config options: mdp.iftype.*.reachable_timeout_ms and interface.*.mdp.reachable_timeout_ms to allow the timeout to be configured. Configure the timeout to 60 seconds in 'routing' tests. Changed the config schema to move options interfaces.*.mdp_tick_ms and interfaces.*.packet_interval into interfaces.*.mdp.tick_ms and interfaces.*.mdp.packet_interval (introduced new sub-struct 'mdp').
2013-10-08 06:06:07 +00:00
SUB_STRUCT(mdp_iftype, mdp,)
New config type: bool_t (char) Replace 'int_boolean' and 'char_boolean' representations with 'boolean'. Clean up spacing and column alignment in conf_schema.h
2013-03-05 02:05:30 +00:00
ATOM(bool_t, send_broadcasts, 1, boolean,, "If false, don't send any broadcast packets")
ATOM(bool_t, default_route, 0, boolean,, "If true, use this interface as a default route")
ATOM(bool_t, prefer_unicast, 0, boolean,, "If true, send unicast data as unicast IP packets if available")
Add per-interace dump of sent and received bytes
2013-07-15 05:57:41 +00:00
ATOM(bool_t, debug, 0, boolean,, "If true, log details of every outgoing packet")
Reduce packet header size for point-to-point links - new interface.x.point_to_point config - disable local echo for point to point dummy interfaces - add ME and YOU sid abbreviation codes after learning other sid
2013-07-16 05:46:07 +00:00
ATOM(bool_t, point_to_point, 0, boolean,, "If true, assume there will only be two devices on this interface")
Add per-interace dump of sent and received bytes
2013-07-15 05:57:41 +00:00
ATOM(bool_t, ctsrts, 0, boolean,, "If true, enable CTS/RTS hardware handshaking")
ATOM(int32_t, uartbps, 57600, int32_rs232baudrate,, "Speed of serial UART link speed (which may be different to serial device link speed)")
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Improve new config prototype code Make config LISTs more flexible. Start implementing existing config options: directory.service (SID parsing), interfaces (better list logic).
2012-11-22 07:50:13 +00:00
Improve config schema declaration Remove array key SORTED flag; all arrays are now sorted by key Use representation's comparison function for sorting arrays, not a schema-supplied comparison function.
2013-03-01 04:57:28 +00:00
ARRAY(interface_list, NO_DUPLICATES)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
KEY_ATOM(unsigned, uint)
VALUE_NODE_STRUCT(network_interface, network_interface)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
END_ARRAY(10)
Improve new config prototype code Make config LISTs more flexible. Start implementing existing config options: directory.service (SID parsing), interfaces (better list logic).
2012-11-22 07:50:13 +00:00
Add remaining config options to new config schema
2012-11-29 02:51:12 +00:00
// The top level.
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
STRUCT(main)
Refactor configuration schema code Replace 'parsefunc' (cf_opt_foo) function arguments in schema declarations with 'repr' (foo) names, which are used to declare cf_opt_foo and cf_fmt_foo.
2013-02-25 06:53:44 +00:00
NODE_STRUCT(interface_list, interfaces, interface_list,)
Improve new config prototype code Refactor and rename, improve logic for reporting unsupported nodes.
2012-11-23 08:43:02 +00:00
SUB_STRUCT(log, log,)
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
SUB_STRUCT(server, server,)
Improve new config prototype code Add monitor.socket and monitor.uid to schema. Remove some deprecated warning functions.
2012-11-28 07:43:53 +00:00
SUB_STRUCT(monitor, monitor,)
Improve new config prototype code Rework ARRAYs, replacing label string with typed key, declaring KEY and VALUE independently. Implement dna.helper.argv array using key of type unsigned short. Implement mdp.iftype.<interfacetype>.tick_ms using key of type short for the interface type code.
2012-11-28 22:54:02 +00:00
SUB_STRUCT(mdp, mdp,)
Improve new config prototype code Refactor ARRAY_xxx() schema declaration macros to fix problem in ARRAY_STRING(). Add dna.helper.executable, dna.helper.argv and server.chdir to config schema.
2012-11-26 07:03:54 +00:00
SUB_STRUCT(dna, dna,)
Overhaul debug flags Replace debugflags_t and DEBUG_XXX bit masks with config schema "debug.xxx" entries. No more support for "debug.all".
2012-12-11 05:29:46 +00:00
SUB_STRUCT(debug, debug,)
Improve new config prototype code Refactor and rename, improve logic for reporting unsupported nodes.
2012-11-23 08:43:02 +00:00
SUB_STRUCT(rhizome, rhizome,)
SUB_STRUCT(directory, directory,)
Improve new config code, SORTED and NO_DUPLICATE flags Config Object Model parser now returns bitmask result of CFxxx flags and only allocates root node if the config file is non-empty. Added emalloc_zero().
2012-11-30 02:02:30 +00:00
SUB_STRUCT(olsr, olsr,)
Improve new config prototype code Add hosts.SIDHEX.{interface,network,port} config options.
2012-11-28 19:58:28 +00:00
SUB_STRUCT(host_list, hosts,)
Improve new config prototype code Refactor to make opt_parsing() functions simpler for ATOM -- no need to check for missing text or unsupported child nodes, or log parse failures. Introduce NODE option which passes whole config_node struct to parsing function.
2012-11-22 23:41:07 +00:00
END_STRUCT
Reference in New Issue Copy Permalink