823 lines
27 KiB
Lua
823 lines
27 KiB
Lua
-- The help text database and query interface.
|
|
--
|
|
-- Help text is read from the rendered text in hack/docs/docs/. If no rendered
|
|
-- text exists, it is read from the script sources (for scripts) or the string
|
|
-- passed to the PluginCommand initializer (for plugins).
|
|
--
|
|
-- There should be one help file for each plugin that contains a summary for the
|
|
-- plugin itself and help for all the commands that plugin provides (if any).
|
|
-- Each script should also have one documentation file.
|
|
--
|
|
-- The database is lazy-loaded when an API method is called. It rechecks its
|
|
-- help sources for updates if an API method has not been called in the last
|
|
-- 60 seconds.
|
|
|
|
local _ENV = mkmodule('helpdb')
|
|
|
|
local argparse = require('argparse')
|
|
|
|
local MAX_STALE_MS = 60000
|
|
|
|
-- paths
|
|
local RENDERED_PATH = 'hack/docs/docs/tools/'
|
|
local TAG_DEFINITIONS = 'hack/docs/docs/Tags.txt'
|
|
|
|
-- used when reading help text embedded in script sources
|
|
local SCRIPT_DOC_BEGIN = '[====['
|
|
local SCRIPT_DOC_END = ']====]'
|
|
local SCRIPT_DOC_BEGIN_RUBY = '=begin'
|
|
local SCRIPT_DOC_END_RUBY = '=end'
|
|
|
|
-- enums
|
|
local ENTRY_TYPES = {
|
|
BUILTIN='builtin',
|
|
PLUGIN='plugin',
|
|
COMMAND='command'
|
|
}
|
|
|
|
local HELP_SOURCES = {
|
|
RENDERED='rendered', -- from the installed, rendered help text
|
|
PLUGIN='plugin', -- from the plugin source code
|
|
SCRIPT='script', -- from the script source code
|
|
STUB='stub', -- from a generated stub
|
|
}
|
|
|
|
-- builtin command names, with aliases mapped to their canonical form
|
|
local BUILTINS = {
|
|
['?']='help',
|
|
alias=true,
|
|
clear='cls',
|
|
cls=true,
|
|
['devel/dump-rpc']=true,
|
|
die=true,
|
|
dir='ls',
|
|
disable=true,
|
|
enable=true,
|
|
fpause=true,
|
|
help=true,
|
|
hide=true,
|
|
keybinding=true,
|
|
['kill-lua']=true,
|
|
['load']=true,
|
|
ls=true,
|
|
man='help',
|
|
plug=true,
|
|
reload=true,
|
|
script=true,
|
|
['sc-script']=true,
|
|
show=true,
|
|
tags=true,
|
|
['type']=true,
|
|
unload=true,
|
|
}
|
|
|
|
---------------------------------------------------------------------------
|
|
-- data structures
|
|
---------------------------------------------------------------------------
|
|
|
|
-- help text database, keys are a subset of the entry database
|
|
-- entry name -> {
|
|
-- help_source (element of HELP_SOURCES),
|
|
-- short_help (string),
|
|
-- long_help (string),
|
|
-- tags (set),
|
|
-- source_timestamp (mtime, 0 for non-files),
|
|
-- source_path (string, nil for non-files)
|
|
-- }
|
|
local textdb = {}
|
|
|
|
-- entry database, points to text in textdb
|
|
-- entry name -> {
|
|
-- entry_types (set of ENTRY_TYPES),
|
|
-- short_help (string, if not nil then overrides short_help in text_entry),
|
|
-- text_entry (string)
|
|
-- }
|
|
--
|
|
-- entry_types is a set because plugin commands can also be the plugin names.
|
|
local entrydb = {}
|
|
|
|
|
|
-- tag name -> list of entry names
|
|
-- Tags defined in the TAG_DEFINITIONS file that have no associated db entries
|
|
-- will have an empty list.
|
|
local tag_index = {}
|
|
|
|
---------------------------------------------------------------------------
|
|
-- data ingestion
|
|
---------------------------------------------------------------------------
|
|
|
|
local function get_rendered_path(entry_name)
|
|
return RENDERED_PATH .. entry_name .. '.txt'
|
|
end
|
|
|
|
local function has_rendered_help(entry_name)
|
|
return dfhack.filesystem.mtime(get_rendered_path(entry_name)) ~= -1
|
|
end
|
|
|
|
local DEFAULT_HELP_TEMPLATE = [[
|
|
%s
|
|
%s
|
|
|
|
No help available.
|
|
]]
|
|
local function make_default_entry(entry_name, help_source, kwargs)
|
|
local default_long_help = DEFAULT_HELP_TEMPLATE:format(
|
|
entry_name, ('='):rep(#entry_name))
|
|
return {
|
|
help_source=help_source,
|
|
short_help='No help available.',
|
|
long_help=default_long_help,
|
|
tags={},
|
|
source_timestamp=kwargs.source_timestamp or 0,
|
|
source_path=kwargs.source_path}
|
|
end
|
|
|
|
-- updates the short_text, the long_text, and the tags in the given entry based
|
|
-- on the text returned from the iterator.
|
|
-- if defined, opts can have the following fields:
|
|
-- begin_marker (string that marks the beginning of the help text; all text
|
|
-- before this marker is ignored)
|
|
-- end_marker (string that marks the end of the help text; text will stop
|
|
-- being parsed after this marker is seen)
|
|
-- no_header (don't try to find the entity name at the top of the help text)
|
|
-- first_line_is_short_help (if set, then read the short help text from the
|
|
-- first commented line of the script instead of
|
|
-- using the first sentence of the long help text.
|
|
-- value is the comment character.)
|
|
local function update_entry(entry, iterator, opts)
|
|
opts = opts or {}
|
|
local lines, tags = {}, ''
|
|
local first_line_is_short_help = opts.first_line_is_short_help
|
|
local begin_marker_found,header_found = not opts.begin_marker,opts.no_header
|
|
local tags_found, short_help_found = false, opts.skip_short_help
|
|
local in_tags, in_short_help = false, false
|
|
for line in iterator do
|
|
if not short_help_found and first_line_is_short_help then
|
|
line = line:trim()
|
|
local _,_,text = line:find('^'..first_line_is_short_help..'%s*(.*)')
|
|
if not text then
|
|
-- if no first-line short help found, fall back to getting the
|
|
-- first sentence of the help text.
|
|
first_line_is_short_help = false
|
|
else
|
|
if not text:endswith('.') then
|
|
text = text .. '.'
|
|
end
|
|
entry.short_help = text
|
|
short_help_found = true
|
|
goto continue
|
|
end
|
|
end
|
|
if not begin_marker_found then
|
|
local _, endpos = line:find(opts.begin_marker, 1, true)
|
|
if endpos == #line then
|
|
begin_marker_found = true
|
|
end
|
|
goto continue
|
|
end
|
|
if opts.end_marker then
|
|
local _, endpos = line:find(opts.end_marker, 1, true)
|
|
if endpos == #line then
|
|
break
|
|
end
|
|
end
|
|
if not header_found and line:find('%w') then
|
|
header_found = true
|
|
elseif in_tags then
|
|
if #line == 0 then
|
|
in_tags = false
|
|
else
|
|
tags = tags .. line
|
|
end
|
|
elseif not tags_found and line:find('^[*]*Tags:[*]*') then
|
|
_,_,tags = line:trim():find('[*]*Tags:[*]* *(.*)')
|
|
in_tags, tags_found = true, true
|
|
elseif not short_help_found and
|
|
line:find('^%s*%w') and not line:find('^%w+:') then
|
|
if in_short_help then
|
|
entry.short_help = entry.short_help .. ' ' .. line
|
|
else
|
|
entry.short_help = line:trim()
|
|
end
|
|
local sentence_end = entry.short_help:find('.', 1, true)
|
|
if sentence_end then
|
|
entry.short_help = entry.short_help:sub(1, sentence_end)
|
|
short_help_found = true
|
|
else
|
|
in_short_help = true
|
|
end
|
|
end
|
|
table.insert(lines, line)
|
|
::continue::
|
|
end
|
|
entry.tags = {}
|
|
for _,tag in ipairs(tags:split('[ ,|]+')) do
|
|
if #tag > 0 and tag_index[tag] then
|
|
entry.tags[tag] = true
|
|
end
|
|
end
|
|
if #lines > 0 then
|
|
entry.long_help = table.concat(lines, '\n')
|
|
end
|
|
end
|
|
|
|
-- create db entry based on parsing sphinx-rendered help text
|
|
local function make_rendered_entry(old_entry, entry_name, kwargs)
|
|
local source_path = get_rendered_path(entry_name)
|
|
local source_timestamp = dfhack.filesystem.mtime(source_path)
|
|
if old_entry and old_entry.help_source == HELP_SOURCES.RENDERED and
|
|
old_entry.source_timestamp >= source_timestamp then
|
|
-- we already have the latest info
|
|
return old_entry
|
|
end
|
|
kwargs.source_path, kwargs.source_timestamp = source_path, source_timestamp
|
|
local entry = make_default_entry(entry_name, HELP_SOURCES.RENDERED, kwargs)
|
|
local ok, lines = pcall(io.lines, source_path)
|
|
if not ok then
|
|
return entry
|
|
end
|
|
update_entry(entry, lines)
|
|
return entry
|
|
end
|
|
|
|
-- create db entry based on the help text in the plugin source (used by
|
|
-- out-of-tree plugins)
|
|
local function make_plugin_entry(old_entry, entry_name, kwargs)
|
|
if old_entry and old_entry.source == HELP_SOURCES.PLUGIN then
|
|
-- we can't tell when a plugin is reloaded, so we can either choose to
|
|
-- always refresh or never refresh. let's go with never for now for
|
|
-- performance.
|
|
return old_entry
|
|
end
|
|
local entry = make_default_entry(entry_name, HELP_SOURCES.PLUGIN, kwargs)
|
|
local long_help = dfhack.internal.getCommandHelp(entry_name)
|
|
if long_help and #long_help:trim() > 0 then
|
|
update_entry(entry, long_help:trim():gmatch('[^\n]*'), {no_header=true})
|
|
end
|
|
return entry
|
|
end
|
|
|
|
-- create db entry based on the help text in the script source (used by
|
|
-- out-of-tree scripts)
|
|
local function make_script_entry(old_entry, entry_name, kwargs)
|
|
local source_path = kwargs.source_path
|
|
local source_timestamp = dfhack.filesystem.mtime(source_path)
|
|
if old_entry and old_entry.source == HELP_SOURCES.SCRIPT and
|
|
old_entry.source_path == source_path and
|
|
old_entry.source_timestamp >= source_timestamp then
|
|
-- we already have the latest info
|
|
return old_entry
|
|
end
|
|
kwargs.source_timestamp, kwargs.entry_type = source_timestamp
|
|
local entry = make_default_entry(entry_name, HELP_SOURCES.SCRIPT, kwargs)
|
|
local ok, lines = pcall(io.lines, source_path)
|
|
if not ok then
|
|
return entry
|
|
end
|
|
local is_rb = source_path:endswith('.rb')
|
|
update_entry(entry, lines,
|
|
{begin_marker=(is_rb and SCRIPT_DOC_BEGIN_RUBY or SCRIPT_DOC_BEGIN),
|
|
end_marker=(is_rb and SCRIPT_DOC_END_RUBY or SCRIPT_DOC_END),
|
|
first_line_is_short_help=(is_rb and '#' or '%-%-')})
|
|
return entry
|
|
end
|
|
|
|
-- updates the dbs (and associated tag index) with a new entry if the entry_name
|
|
-- doesn't already exist in the dbs.
|
|
local function update_db(old_db, entry_name, text_entry, help_source, kwargs)
|
|
if entrydb[entry_name] then
|
|
-- already in db (e.g. from a higher-priority script dir); skip
|
|
return
|
|
end
|
|
entrydb[entry_name] = {
|
|
entry_types=kwargs.entry_types,
|
|
short_help=kwargs.short_help,
|
|
text_entry=text_entry
|
|
}
|
|
if entry_name ~= text_entry then
|
|
return
|
|
end
|
|
|
|
local text_entry, old_entry = nil, old_db[entry_name]
|
|
if help_source == HELP_SOURCES.RENDERED then
|
|
text_entry = make_rendered_entry(old_entry, entry_name, kwargs)
|
|
elseif help_source == HELP_SOURCES.PLUGIN then
|
|
text_entry = make_plugin_entry(old_entry, entry_name, kwargs)
|
|
elseif help_source == HELP_SOURCES.SCRIPT then
|
|
text_entry = make_script_entry(old_entry, entry_name, kwargs)
|
|
elseif help_source == HELP_SOURCES.STUB then
|
|
text_entry = make_default_entry(entry_name, HELP_SOURCES.STUB, kwargs)
|
|
else
|
|
error('unhandled help source: ' .. help_source)
|
|
end
|
|
textdb[entry_name] = text_entry
|
|
end
|
|
|
|
-- add the builtin commands to the db
|
|
local function scan_builtins(old_db)
|
|
local entry_types = {[ENTRY_TYPES.BUILTIN]=true, [ENTRY_TYPES.COMMAND]=true}
|
|
for builtin,canonical in pairs(BUILTINS) do
|
|
if canonical == true then canonical = builtin end
|
|
update_db(old_db, builtin, canonical,
|
|
has_rendered_help(canonical) and
|
|
HELP_SOURCES.RENDERED or HELP_SOURCES.STUB,
|
|
{entry_types=entry_types})
|
|
end
|
|
-- easter egg: replace underline for 'die' help with tombstones
|
|
textdb.die.long_help = textdb.die.long_help:gsub('=', string.char(239))
|
|
end
|
|
|
|
-- scan for enableable plugins and plugin-provided commands and add their help
|
|
-- to the db
|
|
local function scan_plugins(old_db)
|
|
local plugin_names = dfhack.internal.listPlugins()
|
|
for _,plugin in ipairs(plugin_names) do
|
|
local commands = dfhack.internal.listCommands(plugin)
|
|
local includes_plugin = false
|
|
for _,command in ipairs(commands) do
|
|
local kwargs = {entry_types={[ENTRY_TYPES.COMMAND]=true}}
|
|
if command == plugin then
|
|
kwargs.entry_types[ENTRY_TYPES.PLUGIN]=true
|
|
includes_plugin = true
|
|
end
|
|
kwargs.short_help = dfhack.internal.getCommandDescription(command)
|
|
update_db(old_db, command, plugin,
|
|
has_rendered_help(plugin) and
|
|
HELP_SOURCES.RENDERED or HELP_SOURCES.PLUGIN,
|
|
kwargs)
|
|
end
|
|
if not includes_plugin then
|
|
update_db(old_db, plugin, plugin,
|
|
has_rendered_help(plugin) and
|
|
HELP_SOURCES.RENDERED or HELP_SOURCES.STUB,
|
|
{entry_types={[ENTRY_TYPES.PLUGIN]=true}})
|
|
end
|
|
end
|
|
end
|
|
|
|
-- scan for scripts and add their help to the db
|
|
local function scan_scripts(old_db)
|
|
local entry_types = {[ENTRY_TYPES.COMMAND]=true}
|
|
for _,script_path in ipairs(dfhack.internal.getScriptPaths()) do
|
|
local files = dfhack.filesystem.listdir_recursive(
|
|
script_path, nil, false)
|
|
if not files then goto skip_path end
|
|
for _,f in ipairs(files) do
|
|
if f.isdir or
|
|
(not f.path:endswith('.lua') and not f.path:endswith('.rb')) or
|
|
f.path:startswith('test/') or
|
|
f.path:startswith('internal/') then
|
|
goto continue
|
|
end
|
|
local dot_index = f.path:find('%.[^.]*$')
|
|
local entry_name = f.path:sub(1, dot_index - 1)
|
|
local source_path = script_path .. '/' .. f.path
|
|
update_db(old_db, entry_name, entry_name,
|
|
has_rendered_help(entry_name) and
|
|
HELP_SOURCES.RENDERED or HELP_SOURCES.SCRIPT,
|
|
{entry_types=entry_types, source_path=source_path})
|
|
::continue::
|
|
end
|
|
::skip_path::
|
|
end
|
|
end
|
|
|
|
-- read tags and descriptions from the TAG_DEFINITIONS file and add them all
|
|
-- to tag_index, initizlizing each entry with an empty list.
|
|
local function initialize_tags()
|
|
local tag, desc, in_desc = nil, nil, false
|
|
local ok, lines = pcall(io.lines, TAG_DEFINITIONS)
|
|
if not ok then return end
|
|
for line in lines do
|
|
if in_desc then
|
|
line = line:trim()
|
|
if #line == 0 then
|
|
in_desc = false
|
|
goto continue
|
|
end
|
|
desc = desc .. ' ' .. line
|
|
tag_index[tag].description = desc
|
|
else
|
|
_,_,tag,desc = line:find('^%* (%w+)[^:]*: (.+)')
|
|
if not tag then goto continue end
|
|
tag_index[tag] = {description=desc}
|
|
in_desc = true
|
|
end
|
|
::continue::
|
|
end
|
|
end
|
|
|
|
local function index_tags()
|
|
for entry_name,entry in pairs(entrydb) do
|
|
for tag in pairs(textdb[entry.text_entry].tags) do
|
|
-- ignore unknown tags
|
|
if tag_index[tag] then
|
|
table.insert(tag_index[tag], entry_name)
|
|
end
|
|
end
|
|
end
|
|
end
|
|
|
|
-- ensures the db is up to date by scanning all help sources. does not do
|
|
-- anything if it has already been run within the last MAX_STALE_MS milliseconds
|
|
local last_refresh_ms = 0
|
|
local function ensure_db()
|
|
local now_ms = dfhack.getTickCount()
|
|
if now_ms - last_refresh_ms <= MAX_STALE_MS then return end
|
|
last_refresh_ms = now_ms
|
|
|
|
local old_db = textdb
|
|
textdb, entrydb, tag_index = {}, {}, {}
|
|
|
|
initialize_tags()
|
|
scan_builtins(old_db)
|
|
scan_plugins(old_db)
|
|
scan_scripts(old_db)
|
|
index_tags()
|
|
end
|
|
|
|
local function parse_blocks(text)
|
|
local blocks = {}
|
|
for line in text:gmatch('[^\n]*') do
|
|
local _,indent = line:find('^ *')
|
|
table.insert(blocks, {line=line:trim(), indent=indent})
|
|
end
|
|
return blocks
|
|
end
|
|
|
|
local function format_block(line, indent, width)
|
|
local wrapped = line:wrap(width - indent)
|
|
if indent == 0 then return wrapped end
|
|
local padding = (' '):rep(indent)
|
|
local indented_lines = {}
|
|
for line in wrapped:gmatch('[^\n]*') do
|
|
table.insert(indented_lines, padding .. line)
|
|
end
|
|
return table.concat(indented_lines, '\n')
|
|
end
|
|
|
|
-- wraps the unwrapped source help at the specified width, preserving block
|
|
-- indents
|
|
local function rewrap(text, width)
|
|
local formatted_blocks = {}
|
|
for _,block in ipairs(parse_blocks(text)) do
|
|
table.insert(formatted_blocks,
|
|
format_block(block.line, block.indent, width))
|
|
end
|
|
return table.concat(formatted_blocks, '\n')
|
|
end
|
|
|
|
---------------------------------------------------------------------------
|
|
-- get API
|
|
---------------------------------------------------------------------------
|
|
|
|
-- converts strings into single-element lists containing that string
|
|
local function normalize_string_list(l)
|
|
if not l or #l == 0 then return nil end
|
|
if type(l) == 'string' then
|
|
return {l}
|
|
end
|
|
return l
|
|
end
|
|
|
|
local function has_keys(str, dict)
|
|
if not str or #str == 0 then
|
|
return false
|
|
end
|
|
for _,s in ipairs(normalize_string_list(str)) do
|
|
if not dict[s] then
|
|
return false
|
|
end
|
|
end
|
|
return true
|
|
end
|
|
|
|
-- returns whether the given string (or list of strings) is an entry (are all
|
|
-- entries) in the db
|
|
function is_entry(str)
|
|
ensure_db()
|
|
return has_keys(str, entrydb)
|
|
end
|
|
|
|
local function get_db_property(entry_name, property)
|
|
ensure_db()
|
|
if not entrydb[entry_name] then
|
|
error(('helpdb entry not found: "%s"'):format(entry_name))
|
|
end
|
|
return entrydb[entry_name][property] or
|
|
textdb[entrydb[entry_name].text_entry][property]
|
|
end
|
|
|
|
function get_entry_types(entry)
|
|
return get_db_property(entry, 'entry_types')
|
|
end
|
|
|
|
-- returns the ~54 char summary blurb associated with the entry
|
|
function get_entry_short_help(entry)
|
|
return get_db_property(entry, 'short_help')
|
|
end
|
|
|
|
-- returns the full help documentation associated with the entry, optionally
|
|
-- wrapped to the specified width (80 if not specified).
|
|
function get_entry_long_help(entry, width)
|
|
return rewrap(get_db_property(entry, 'long_help'), width or 80)
|
|
end
|
|
|
|
-- returns the set of tags associated with the entry
|
|
function get_entry_tags(entry)
|
|
return get_db_property(entry, 'tags')
|
|
end
|
|
|
|
-- returns whether the given string (or list of strings) matches a tag name
|
|
function is_tag(str)
|
|
ensure_db()
|
|
return has_keys(str, tag_index)
|
|
end
|
|
|
|
local function set_to_sorted_list(set)
|
|
local list = {}
|
|
for item in pairs(set) do
|
|
table.insert(list, item)
|
|
end
|
|
table.sort(list)
|
|
return list
|
|
end
|
|
|
|
-- returns the defined tags in alphabetical order
|
|
function get_tags()
|
|
ensure_db()
|
|
return set_to_sorted_list(tag_index)
|
|
end
|
|
|
|
function get_tag_data(tag)
|
|
ensure_db()
|
|
if not tag_index[tag] then
|
|
error(('helpdb tag not found: "%s"'):format(tag))
|
|
end
|
|
return tag_index[tag]
|
|
end
|
|
|
|
---------------------------------------------------------------------------
|
|
-- search API
|
|
---------------------------------------------------------------------------
|
|
|
|
-- returns a list of path elements in reverse order
|
|
local function chunk_for_sorting(str)
|
|
local parts = str:split('/')
|
|
local chunks = {}
|
|
for i=1,#parts do
|
|
chunks[#parts - i + 1] = parts[i]
|
|
end
|
|
return chunks
|
|
end
|
|
|
|
-- sorts by last path component, then by parent path components.
|
|
-- something comes before nothing.
|
|
-- e.g. gui/autofarm comes immediately before autofarm
|
|
function sort_by_basename(a, b)
|
|
local a = chunk_for_sorting(a)
|
|
local b = chunk_for_sorting(b)
|
|
local i = 1
|
|
while a[i] do
|
|
if not b[i] then
|
|
return true
|
|
end
|
|
if a[i] ~= b[i] then
|
|
return a[i] < b[i]
|
|
end
|
|
i = i + 1
|
|
end
|
|
return false
|
|
end
|
|
|
|
-- returns true if all filter elements are matched (i.e. any of the tags AND
|
|
-- any of the strings AND any of the entry_types)
|
|
local function matches(entry_name, filter)
|
|
if filter.tag then
|
|
local matched = false
|
|
local tags = get_db_property(entry_name, 'tags')
|
|
for _,tag in ipairs(filter.tag) do
|
|
if tags[tag] then
|
|
matched = true
|
|
break
|
|
end
|
|
end
|
|
if not matched then
|
|
return false
|
|
end
|
|
end
|
|
if filter.entry_type then
|
|
local matched = false
|
|
local etypes = get_db_property(entry_name, 'entry_types')
|
|
for _,etype in ipairs(filter.entry_type) do
|
|
if etypes[etype] then
|
|
matched = true
|
|
break
|
|
end
|
|
end
|
|
if not matched then
|
|
return false
|
|
end
|
|
end
|
|
if filter.str then
|
|
local matched = false
|
|
for _,str in ipairs(filter.str) do
|
|
if entry_name:find(str, 1, true) then
|
|
matched = true
|
|
break
|
|
end
|
|
end
|
|
if not matched then
|
|
return false
|
|
end
|
|
end
|
|
return true
|
|
end
|
|
|
|
local function matches_any(entry_name, filters)
|
|
for _,filter in ipairs(filters) do
|
|
if matches(entry_name, filter) then
|
|
return true
|
|
end
|
|
end
|
|
return false
|
|
end
|
|
|
|
-- normalizes the lists in the filter and returns nil if no filter elements are
|
|
-- populated
|
|
local function normalize_filter_map(f)
|
|
if not f then return nil end
|
|
local filter = {}
|
|
filter.str = normalize_string_list(f.str)
|
|
filter.tag = normalize_string_list(f.tag)
|
|
filter.entry_type = normalize_string_list(f.entry_type)
|
|
if not filter.str and not filter.tag and not filter.entry_type then
|
|
return nil
|
|
end
|
|
return filter
|
|
end
|
|
|
|
local function normalize_filter_list(fs)
|
|
if not fs then return nil end
|
|
local filter_list = {}
|
|
for _,f in ipairs(#fs > 0 and fs or {fs}) do
|
|
table.insert(filter_list, normalize_filter_map(f))
|
|
end
|
|
if #filter_list == 0 then return nil end
|
|
return filter_list
|
|
end
|
|
|
|
-- returns a list of entry names, alphabetized by their last path component,
|
|
-- with populated path components coming before null path components (e.g.
|
|
-- autobutcher will immediately follow gui/autobutcher).
|
|
-- the optional include and exclude filter params are maps (or lists of maps)
|
|
-- with the following elements:
|
|
-- str - if a string, filters by the given substring. if a table of strings,
|
|
-- includes entry names that match any of the given substrings.
|
|
-- tag - if a string, filters by the given tag name. if a table of strings,
|
|
-- includes entries that match any of the given tags.
|
|
-- entry_type - if a string, matches entries of the given type. if a table of
|
|
-- strings, includes entries that match any of the given types. valid
|
|
-- types are: "builtin", "plugin", "command". note that many plugin
|
|
-- commands have the same name as the plugin, so those entries will
|
|
-- match both "plugin" and "command" types.
|
|
-- filter elements in a map are ANDed together (e.g. if both str and tag are
|
|
-- specified, the match is on any of the str elements AND any of the tag
|
|
-- elements). If lists of maps are passed, the maps are ORed (that is, the match
|
|
-- succeeds if any of the filters match).
|
|
function search_entries(include, exclude)
|
|
ensure_db()
|
|
include = normalize_filter_list(include)
|
|
exclude = normalize_filter_list(exclude)
|
|
local entries = {}
|
|
for entry in pairs(entrydb) do
|
|
if (not include or matches_any(entry, include)) and
|
|
(not exclude or not matches_any(entry, exclude)) then
|
|
table.insert(entries, entry)
|
|
end
|
|
end
|
|
table.sort(entries, sort_by_basename)
|
|
return entries
|
|
end
|
|
|
|
-- returns a list of all commands. used by Core's autocomplete functionality.
|
|
function get_commands()
|
|
local include = {entry_type=ENTRY_TYPES.COMMAND}
|
|
return search_entries(include)
|
|
end
|
|
|
|
function is_builtin(command)
|
|
return is_entry(command) and get_entry_types(command)[ENTRY_TYPES.BUILTIN]
|
|
end
|
|
|
|
---------------------------------------------------------------------------
|
|
-- print API (outputs to console)
|
|
---------------------------------------------------------------------------
|
|
|
|
-- implements the 'help' builtin command
|
|
function help(entry)
|
|
ensure_db()
|
|
if not entrydb[entry] then
|
|
dfhack.printerr(('No help entry found for "%s"'):format(entry))
|
|
return
|
|
end
|
|
print(get_entry_long_help(entry))
|
|
end
|
|
|
|
-- prints col1text (width 21), a one space gap, and col2 (width 58)
|
|
-- if col1text is longer than 21 characters, col2text is printed starting on the
|
|
-- next line. if col2text is longer than 58 characters, it is wrapped. col2text
|
|
-- lines on lines below the col1text output are indented by one space further
|
|
-- than the col2text on the first line.
|
|
local COL1WIDTH, COL2WIDTH = 20, 58
|
|
local function print_columns(col1text, col2text)
|
|
col2text = col2text:wrap(COL2WIDTH)
|
|
local wrapped_col2 = {}
|
|
for line in col2text:gmatch('[^'..NEWLINE..']*') do
|
|
table.insert(wrapped_col2, line)
|
|
end
|
|
local col2_start_line = 1
|
|
if #col1text > COL1WIDTH then
|
|
print(col1text)
|
|
else
|
|
print(('%-'..COL1WIDTH..'s %s'):format(col1text, wrapped_col2[1]))
|
|
col2_start_line = 2
|
|
end
|
|
for i=col2_start_line,#wrapped_col2 do
|
|
print(('%'..COL1WIDTH..'s %s'):format(' ', wrapped_col2[i]))
|
|
end
|
|
end
|
|
|
|
-- prints the requested entries to the console. include and exclude filters are
|
|
-- defined as in search_entries() above.
|
|
local function list_entries(skip_tags, include, exclude)
|
|
local entries = search_entries(include, exclude)
|
|
for _,entry in ipairs(entries) do
|
|
local short_help = get_entry_short_help(entry)
|
|
if not skip_tags then
|
|
local tags = set_to_sorted_list(get_entry_tags(entry))
|
|
if #tags > 0 then
|
|
local taglist = table.concat(tags, ', ')
|
|
short_help = short_help .. NEWLINE .. 'tags: ' .. taglist
|
|
end
|
|
end
|
|
print_columns(entry, short_help)
|
|
end
|
|
if #entries == 0 then
|
|
print('No matches.')
|
|
end
|
|
end
|
|
|
|
-- wraps the list_entries() API to provide a more convenient interface for Core
|
|
-- to implement the 'ls' builtin command.
|
|
-- filter_str - if a tag name (or a list of tag names), will filter by that
|
|
-- tag/those tags. otherwise, will filter as a substring/list of
|
|
-- substrings
|
|
-- skip_tags - whether to skip printing tag info
|
|
-- show_dev_commands - if true, will include scripts in the modtools/ and
|
|
-- devel/ directories. otherwise those scripts will be
|
|
-- excluded
|
|
-- exclude_strs - comma-separated list of strings. entries are excluded if
|
|
-- they match any of the strings.
|
|
function ls(filter_str, skip_tags, show_dev_commands, exclude_strs)
|
|
local include = {entry_type={ENTRY_TYPES.COMMAND}}
|
|
if is_tag(filter_str) then
|
|
include.tag = filter_str
|
|
else
|
|
include.str = filter_str
|
|
end
|
|
local excludes = {}
|
|
if exclude_strs and #exclude_strs > 0 then
|
|
table.insert(excludes, {str=argparse.stringList(exclude_strs)})
|
|
end
|
|
if not show_dev_commands then
|
|
table.insert(excludes, {tag='dev'})
|
|
end
|
|
list_entries(skip_tags, include, excludes)
|
|
end
|
|
|
|
local function list_tags()
|
|
local tags = get_tags()
|
|
for _,tag in ipairs(tags) do
|
|
print_columns(tag, get_tag_data(tag).description)
|
|
end
|
|
end
|
|
|
|
-- implements the 'tags' builtin command
|
|
function tags(tag)
|
|
if tag and #tag > 0 and not is_tag(tag) then
|
|
dfhack.printerr(('unrecognized tag: "%s"'):format(tag))
|
|
end
|
|
|
|
if not is_tag(tag) then
|
|
list_tags()
|
|
return
|
|
end
|
|
|
|
local skip_tags = true
|
|
local include = {entry_type={ENTRY_TYPES.COMMAND}, tag=tag}
|
|
list_entries(skip_tags, include)
|
|
end
|
|
|
|
return _ENV
|