README

ilib-loctool-po

An ilib loctool plugin to parse and localize GNU gettext po files.

This plugin can parse and localize po or pot files written by tools such as easygettext.

Regular PO and POT Files

Entries in a PO file have the following format:

white-space
#  translator-comments
#. extracted-comments
#: reference…
#, flag…
#| msgctxt previous-context
#| msgid previous-untranslated-string
msgctxt context
msgid untranslated-string
msgstr translated-string

The format is specified on the gnu website.

This plugin can read any file with this format. Individual entries are registered at string resources and will appear in the extracted xliff and the new strings files.

If an entry has a context line, it will get a context in the xliff output and will be differentiated from other strings that have the same source but a different (or missing). That is, two entries that differ only by their context will appear as two separate strings in the xliff output.

In POT files, there are often no translations, as it is the "template" that is used to generate all the translated po files. This plugin can support any string with or without translations.

Plurals

This plugin can also support plural strings. Plurals have the following format:

msgid untranslated-string-singular
msgid_plural untranslated-string-plural
msgstr[0] translated-string-case-0
...
msgstr[N] translated-string-case-n

The string in msgid will be considered to be the string for the one category used as the singular in English source. The msgstr_plural will be considered the other category, which is used as the plural string in English. See the Unicode plural categories document for details on what the categories mean.

The lines msgstr[0] through msgstr[N] are the translations of these categories. The numbers are mapped to Unicode categories according to the mappings included below in Appendix A. The mappings are different for each language. That is, in one language, 0 might map to the one category, whereas in another, it might be the zero category. The strings in the xliff output will include the category only, and the categories will be mapped back to numbers in the translated po file output.

Note that for some languages, the number of categories is different than in English. For example, English uses 2 categories (one and other) but Russian uses 3 (one, few, and other). The xliff output will contain the right number of category strings for the target language. Your translators should note the category given in the extype attribute when translating each string. They will not have to add or subtract any xliff trans-unit entries. They only need to translate the contents of the target tag in the given trans-units.

Example

Input po or pot file:

msgid "There is {n} object."
msgstr_plural "There are {n} objects."

Output xliff snippet:

      <trans-unit id="4" resname="There is {n} object." restype="plural" datatype="po" extype="one">
        <source>There is {n} object.</source>
        <target state="new">There is {n} object.</target>
      </trans-unit>
      <trans-unit id="5" resname="There is {n} object." restype="plural" datatype="po" extype="few">
        <source>There are {n} objects.</source>
        <target state="new">There are {n} objects.</target>
      </trans-unit>
      <trans-unit id="6" resname="There is {n} object." restype="plural" datatype="po" extype="other">
        <source>There are {n} objects.</source>
        <target state="new">There are {n} objects.</target>
      </trans-unit>

Note that the msgid string (the English singular) is used as the resource name (resname) for each of the xliff entries so that they can be connected again later when the xliff file is read in. Also, the source string for every category other than one is the source string given on the msgstr_plural line.

Configuring the Plugin

The plugin will look for the po property within the settings of your project.json file. The following setting is used within the po property:

mappings: a mapping between file matchers and an object that gives info used to localize the files that match it. This allows different po files within the project to be processed into different output files. The matchers are a micromatch-style string, similar to the the includes and excludes section of a project.json file. The value of that mapping is an object that can contain the following properties:
- template: a path template to use to generate the path to the translated output files. The template replaces strings in square brackets with special values, and keeps any characters intact that are not in square brackets. The default template, if not specified is "[dir]/[locale].po". The plugin recognizes and replaces the following strings in template strings:
  - [dir] the original directory where the matched source file came from. This is given as a directory that is relative to the root of the project. eg. "foo/bar/strings.po" -> "foo/bar"
  - [filename] the file name of the matching file. eg. "foo/bar/strings.po" -> "strings.po"
  - [basename] the basename of the matching file without any extension eg. "foo/bar/strings.po" -> "strings"
  - [extension] the extension part of the file name of the source file. etc. "foo/bar/strings.po" -> "po"
  - [locale] the full BCP-47 locale specification for the target locale eg. "zh-Hans-CN" -> "zh-Hans-CN"
  - [language] the language portion of the full locale eg. "zh-Hans-CN" -> "zh"
  - [script] the script portion of the full locale eg. "zh-Hans-CN" -> "Hans"
  - [region] the region portion of the full locale eg. "zh-Hans-CN" -> "CN"
  - [localeDir] the full locale where each portion of the locale is a directory in this order: [langage], [script], [region]. eg, "zh-Hans-CN" -> "zh/Hans/CN", but "en" -> "en".
  - [localeUnder] the full BCP-47 locale specification, but using underscores to separate the locale parts instead of dashes. eg. "zh-Hans-CN" -> "zh_Hans_CN"
- localeMap: an output locale map that maps the locale used in the translations to a different locale for use in the file name template. For example, an app may wish to use the locale specifier "zh-CN" instead of the full "zh-Hans-CN" for some output files, but not all of them. In this case, a locale map for this template mapping is how this can be achieved. Any locale not listed in the mapping will be used as-is. The overall [shared] locale map is also applied if there is no locale map in the template mapping for a particular locale.
- ignoreComments: specify whether to ignore types of comments. Possible values are:
  - true All comment types should be ignored
  - false No comment types should be ignored (default)
  - Array of strings. Name the types of comments that should be ignored. Possible values are:
    - "translator" - ignore translator comments (prefix is "# ")
    - "extracted" - ignore comments extracted from the source code (prefix is "#.")
    - "flags" - ignore special processing flags (prefix is "#,")
    - "previous" - ignore previous translation (prefix is "#|")
    - "paths" - ignore file names and line numbers (prefix is "#:")
- headerLocale: specify what kind of locale to put in the file header. Possible values are:
  - full: put the fully specified locale spec in the header
  - abbreviated: put an abbreviated locale spec in the header (language only)
  - mapped: put the results of the output locale mapping (see above) into the header
- contextInKey: some translation management systems cannot support separate context fields, and therefore two translation units that only differ in their context cannot be distinguished from each other. If this setting is set to "true", then the context is added to the key for a unit. eg. if the string is "Sent" and the context is "Email", a translation unit will be produced with the key "Sent --- Email" and the source string is "Sent".

Example configuration:

{
    "settings": {
        "po": {
            "mappings": {
                "**/template.pot": {
                    "template": "resources/[locale].po",
                    "ignoreComments": ["paths"]
                },
                "sublibrary/**/library.pot": {
                    "template": "[dir]/[locale].po",
                    "localeMap": {
                        "nb-NO": "no",
                        "sr-Cyrl-RS": "sr-RS",
                        "zh-Hans-CN": "zh-CN"
                    },
                    "ignoreComments": true
                }
            }
        }
    }
}

In the above example, any file named template.pot will be localized and the output is sent to a file named after the target locale located in the resources directory. In these files, the translator and extracted type of comments are ignored and will not be extracted into the resources and therefore will not appear in the xliff files.

In the second part of the example, any library.pot file that appears in the sublibrary directory will be localized and the results sent to a po file named after each target locale which will appear in the same directory where the source file was located. If the locale is one of the ones listed in the locale map, it will be mapped before being substituted in to the template. For these files, all comment types are ignored and none are sent to the translators in the xliff files.

If the name of the localized file that the template produces is the same as the source file name, this plugin will throw an exception, the file will not be localized, and the loctool will continue on to the next file.

License

This plugin is license under Apache2. See the LICENSE file for more details.

Release Notes

v1.6.1

make sure the msgctxt line comes out before the msgid line. Some libraries cannot handle it the other way around.

v1.6.0

added the contextInKey setting so that we can support translation management systems that do not support contexts. Instead, we add the context to the key for each translation unit, so that two translations units that differ only in their context can still be dinstinguished from each other.

v1.5.1

fixed a bug in the getOutputLocale function when you are missing a locale in the localeMap that is mentioned in the list of locales

v1.5.0

Added headerLocale setting to the mappings. This allows you to specify the style of the locale spec listed in the header of the po file as it is written out to disk.

v1.4.1

Fixed the po header output so that it shows the locale spec of the mapped output locale, not of the source locale.

v1.4.0

added the ability to ignore comments. This solves the problem where file names and line numbers for each resource change when someone makes an unrelated change to a source file, but the resource itself nor the code around it has changed. This causes some translation management systems to treat the string as changed and therefore requiring useless retranslation.
- Added the "ignoreComments" config option (see above for details)

v1.3.1

Fixed a bug where the plugin would not work when generating pseudo localizations for plurals

v1.3.0

Added the ability to specify a locale map in the file name template mapping. This allows for different locale specifications for different sets of output files, which may be necessary if those output files are intended to be used in different programming languages or on different platforms that support a different set of locales.

v1.2.2

Fix a bug where the target locale was not used when specified to POFileType.newFile. It was never passed in to the POFile constructor.
Fix a bug where the target locale was specified for string and plural resources and there were no target strings or plurals. Having the locale but no target confuses mojito.

v1.2.1

Make sure to output an empty string if the translated string is the same as the source string so that gettext() will default back to the source string

v1.2.0

added the ability to specify a locale map so that output file names of po files are mapped from the internal locale to the output locale

v1.1.2

fixed a bug where path names in #: comments that did not have a colon and a line number were not being extracted properly

v1.1.1

fixed a bug where every resource from the PO file had its own file tag in the xliff output because the "original" path was set to the file name colon line number.

v1.1.0

Added the ability to use po files as output resource files by adding a write method. This means it can also be used as an output format for the new convert action.
- if resources are added where the target locale does not match the locale of the PO file, then those resources will be added as source-only resources
- handles resources with missing translations and puts a placeholder entry into the PO file
- handles missing plural strings as well, depending on the language

v1.0.1

Make this plugin able to read already-localized po files The output file name template is used to construct a regular expression to recognize already localized files and what the locale of that file is. Without the template, the locale was never extracted and the source and target were both en-US which is not correct. This was a bigger problem for those languages where the plural resources have more plural categories than in English, such as Russian or Polish.

v1.0.0

initial version
read regular po and pot files and output translated po files

Appendix A. Mappings From GNU gettext Plural Number and Unicode Category

These mappings are encoded in the file pluralforms.json which is packaged with this plugin. The languages are listed below by their ISO 639 2 letter codes. This plugin assumes that any language not explicitly listed below has only the one and other categories.

ar

number	category
0	zero
1	one
2	two
3	few
4	many
5	other

be

number	category
0	one
1	few
2	other

bg

number	category
0	one
1	other

cs

number	category
0	one
1	few
2	other

da

number	category
0	one
1	other

de

number	category
0	one
1	other

el

number	category
0	one
1	other

en

number	category
0	one
1	other

eo

number	category
0	one
1	other

es

number	category
0	one
1	other

et

number	category
0	one
1	other

fi

number	category
0	one
1	other

fo

number	category
0	one
1	other

fr

number	category
0	one
1	other

ga

number	category
0	one
1	few
2	other

he

number	category
0	one
1	other

hr

number	category
0	one
1	few
2	other

hu

number	category
0	one
1	other

id

number	category
0	one
1	other

it

number	category
0	one
1	other

ja

number	category
0	other

ko

number	category
0	other

lt

number	category
0	one
1	few
2	other

lv

number	category
0	zero
1	one
2	other

nl

number	category
0	one
1	other

no

number	category
0	one
1	other

pl

number	category
0	one
1	few
2	other

pt

number	category
0	one
1	other

ro

number	category
0	one
1	few
2	other

ru

number	category
0	one
1	few
2	other

sk

number	category
0	one
1	few
2	other

sl

number	category
0	one
1	two
2	few
3	other

sr

number	category
0	one
1	few
2	other

sv

number	category
0	one
1	other

th

number	category
0	other

tr

number	category
0	one
1	other

uk

number	category
0	one
1	few
2	other

zh

number	category
0	other

Usage no npm install needed!