⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 gettext.jam

📁 Boost provides free peer-reviewed portable C++ source libraries. We emphasize libraries that work
💻 JAM
字号:
# Copyright 2003, 2004, 2005, 2006 Vladimir Prus# Distributed under the Boost Software License, Version 1.0.# (See accompanying file LICENSE_1_0.txt or http://www.boost.org/LICENSE_1_0.txt)#  This module support GNU gettext internationalization utilities.##  It provides two main target rules: 'gettext.catalog', used for#  creating machine-readable catalogs from translations files, and#  'gettext.update', used for update translation files from modified#  sources.##  To add i18n support to your application you should follow these#  steps.##  - Decide on a file name which will contain translations and#  what main target name will be used to update it. For example::##    gettext.update update-russian : russian.po a.cpp my_app ;##  - Create the initial translation file by running::##    bjam update-russian##  - Edit russian.po. For example, you might change fields like LastTranslator.##  - Create a main target for final message catalog::##    gettext.catalog russian : russian.po ;##  The machine-readable catalog will be updated whenever you update#  "russian.po". The "russian.po" file will be updated only on explicit#  request. When you're ready to update translations, you should##  - Run::##    bjam update-russian##  - Edit "russian.po" in appropriate editor.##  The next bjam run will convert "russian.po" into machine-readable form.##  By default, translations are marked by 'i18n' call. The 'gettext.keyword'#  feature can be used to alter this.import targets ;import property-set ;import virtual-target ;import "class" : new ;import project ;import type ;import generators ;import errors ;import feature : feature ;import toolset : flags ;import regex ;.path = "" ;# Initializes the gettext module.rule init ( path ? # Path where all tools are located. If not specified,                   # they should be in PATH.          ){    if $(.initialized) && $(.path) != $(path)    {        errors.error "Attempt to reconfigure with different path" ;    }    .initialized = true ;    if $(path)    {        .path = $(path)/ ;    }}# Creates a main target 'name', which, when updated, will cause# file 'existing-translation' to be updated with translations# extracted from 'sources'. It's possible to specify main target# in sources --- it which case all target from dependency graph# of those main targets will be scanned, provided they are of# appropricate type. The 'gettext.types' feature can be used to# control the types.## The target will be updated only if explicitly requested on the# command line.rule update ( name : existing-translation sources + : requirements * ){    local project = [ project.current ] ;    targets.main-target-alternative      [ new typed-target $(name) : $(project) : gettext.UPDATE :        $(existing-translation) $(sources)        : [ targets.main-target-requirements $(requirements) : $(project) ]      ] ;    $(project).mark-target-as-explicit $(name) ;}# The human editable source, containing translation.type.register gettext.PO : po ;# The machine readable message catalog.type.register gettext.catalog : mo ;# Intermediate type produce by extracting translations from# sources.type.register gettext.POT : pot ;# Pseudo type used to invoke update-translations generatortype.register gettext.UPDATE ;# Identifies the keyword that should be used when scanning sources.# Default: i18nfeature gettext.keyword : : free ;# Contains space-separated list of sources types which should be scanned.# Default: "C CPP"feature gettext.types : : free ;generators.register-standard gettext.compile : gettext.PO : gettext.catalog ;class update-translations-generator : generator{    import regex : split ;    import property-set ;    rule __init__ ( * : * )    {        generator.__init__ $(1) : $(2) : $(3) : $(4) : $(5) : $(6) : $(7) : $(8) : $(9) ;    }    # The rule should be called with at least two sources. The first source    # is the translation (.po) file to update. The remaining sources are targets    # which should be scanned for new messages. All sources files for those targets    # will be found and passed to the 'xgettext' utility, which extracts the    # messages for localization. Those messages will be merged to the .po file.    rule run ( project name ? : property-set : sources * : multiple ? )    {        local types = [ $(property-set).get <gettext.types> ] ;        types ?= "C CPP" ;        types = [ regex.split $(types) " " ] ;        local keywords = [ $(property-set).get <gettext.keyword> ] ;        property-set = [ property-set.create $(keywords:G=<gettext.keyword>) ] ;        # First deterime the list of sources that must be scanned for        # messages.        local all-sources ;        # CONSIDER: I'm not sure if the logic should be the same as for 'stage':        # i.e. following dependency properties as well.        for local s in $(sources[2-])        {            all-sources += [ virtual-target.traverse $(s) : : include-sources ] ;        }        local right-sources ;        for local s in $(all-sources)        {            if [ $(s).type ] in $(types)            {                right-sources += $(s) ;            }        }        local .constructed ;        if $(right-sources)        {            # Create the POT file, which will contain list of messages extracted            # from the sources.            local extract =              [ new action $(right-sources) : gettext.extract : $(property-set) ] ;            local new-messages = [ new file-target $(name) : gettext.POT              : $(project) : $(extract) ] ;            # Create a notfile target which will update the existing translation file            # with new messages.            local a = [ new action $(sources[1]) $(new-messages)              : gettext.update-po-dispatch ] ;            local r = [ new notfile-target $(name) : $(project) : $(a) ] ;            .constructed = [ virtual-target.register $(r) ] ;        }        else        {            errors.error "No source could be scanned by gettext tools" ;        }        return $(.constructed) ;    }}generators.register [ new update-translations-generator gettext.update : : gettext.UPDATE ] ;flags gettext.extract KEYWORD <gettext.keyword> ;actions extract{    $(.path)xgettext -k$(KEYWORD:E=i18n) -o $(<) $(>)}# Does realy updating of po file. The tricky part is that# we're actually updating one of the sources:# $(<) is the NOTFILE target we're updating# $(>[1]) is the PO file to be really updated.# $(>[2]) is the PO file created from sources.## When file to be updated does not exist (during the# first run), we need to copy the file created from sources.# In all other cases, we need to update the file.rule update-po-dispatch{    NOCARE $(>[1]) ;    gettext.create-po $(<) : $(>) ;    gettext.update-po $(<) : $(>) ;    _ on $(<) = " " ;    ok on $(<) = "" ;    EXISTING_PO on $(<) = $(>[1]) ;}# Due to fancy interaction of existing and updated, this rule can be called with# one source, in which case we copy the lonely source into EXISTING_PO, or with# two sources, in which case the action body expands to nothing. I'd really like# to have "missing" action modifier.actions quietly existing updated create-po bind EXISTING_PO{    cp$(_)"$(>[1])"$(_)"$(EXISTING_PO)"$($(>[2]:E=ok))}actions updated update-po bind EXISTING_PO{    $(.path)msgmerge$(_)-U$(_)"$(EXISTING_PO)"$(_)"$(>[1])"}actions gettext.compile{    $(.path)msgfmt -o $(<) $(>)}IMPORT $(__name__) : update : : gettext.update ;

⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -