If this tool saves you time, please consider making a donation towards the continued maintainence and development: https://donate.biesbjerg.com
Extract translatable (ngx-translate) strings and save as a JSON or Gettext pot file. Merges with existing strings if the output file already exists.
Install the package in your project:
npm install @biesbjerg/ngx-translate-extract --save-dev
Add a script to your project's package.json
:
...
"scripts": {
"i18n:init": "ngx-translate-extract --input ./src --output ./src/assets/i18n/template.json --key-as-default-value --replace --format json",
"i18n:extract": "ngx-translate-extract --input ./src --output ./src/assets/i18n/{en,da,de,fi,nb,nl,sv}.json --clean --format json"
}
...
You can now run npm run i18n:extract
and it will extract strings from your project.
Extract from dir and save to file
ngx-translate-extract --input ./src --output ./src/assets/i18n/strings.json
Extract from multiple dirs
ngx-translate-extract --input ./src-a ./src-b --output ./src/assets/i18n/strings.json
Extract and save to multiple files using path expansion
ngx-translate-extract --input ./src --output ./src/i18n/{da,en}.json
Tabs are used by default for indentation when saving extracted strings in json formats:
If you want to use spaces instead, you can do the following:
ngx-translate-extract --input ./src --output ./src/i18n/en.json --format-indentation ' '
If you want to extract strings that are not passed directly to TranslateService
's get()
/instant()
/stream()
methods, you can wrap them in a marker function to let ngx-translate-extract
know you want to extract them.
Install marker function:
npm install @biesbjerg/ngx-translate-extract-marker
import { marker } from '@biesbjerg/ngx-translate-extract-marker';
marker('Extract me');
You can alias the marker function if needed:
import { marker as _ } from '@biesbjerg/ngx-translate-extract-marker';
_('Extract me');
Note: ngx-translate-extract
will automatically detect the import name
Usage:
ngx-translate-extract [options]
Output
--format, -f Format [string] [choices: "json", "namespaced-json", "pot"] [default: "json"]
--format-indentation, --fi Format indentation (JSON/Namedspaced JSON) [string] [default: " "]
--sort, -s Sort strings in alphabetical order [boolean]
--clean, -c Remove obsolete strings after merge [boolean]
--replace, -r Replace the contents of output file if it exists (Merges by default) [boolean]
Extracted key value (defaults to empty string)
--key-as-default-value, -k Use key as default value [boolean]
--null-as-default-value, -n Use null as default value [boolean]
--string-as-default-value, -d Use string as default value [string]
Options:
--version, -v Show version number [boolean]
--help, -h Show help [boolean]
--input, -i Paths you would like to extract strings from. You can use path expansion, glob patterns and
multiple paths [array] [required] [default: ["/Users/kim/apps/ngx-translate-extract"]]
--output, -o Paths where you would like to save extracted strings. You can use path expansion, glob
patterns and multiple paths [array] [required]
Examples:
ngx-translate-extract -i ./src-a/ -i ./src-b/ -o strings.json Extract (ts, html) from multiple paths
ngx-translate-extract -i './{src-a,src-b}/' -o strings.json Extract (ts, html) from multiple paths using brace
expansion
ngx-translate-extract -i ./src/ -o ./i18n/da.json -o ./i18n/en.json Extract (ts, html) and save to da.json and en.json
ngx-translate-extract -i ./src/ -o './i18n/{en,da}.json' Extract (ts, html) and save to da.json and en.json
using brace expansion
ngx-translate-extract -i './src/**/*.{ts,tsx,html}' -o strings.json Extract from ts, tsx and html
ngx-translate-extract -i './src/**/!(*.spec).{ts,html}' -o Extract from ts, html, excluding files with ".spec"
strings.json
Please pay attention of which version of gettext-parser
you actually use in your project. For instance, gettext-parser:1.2.2
does not support HTML tags in translation keys.