Sometimes, we want to merge an initial version of a feature without deciding on the final text that users will see. This issue is to figure out a process for the central-frontend repo to facilitate that. The goals are:
- Don't send messages to Transifex that are still preliminary. We've seen cases where translators translate a message that we ultimately replace before release.
- Make it easy to identify placeholder text so that we can replace it and also check for its absence prior to release.
Here are a couple of ideas for how to implement this:
Idea 1. Just specify raw text in the <template>, followed by a word like TODO. We could search the codebase for TODO.
Idea 2. Continue to use $t() in the <template>, but make it possible to add messages to <i18n> blocks without sending them to Transifex. One way to do this would be via magic comments, which our Transifex scripts already know how to identify. Something like:
{
"en": {
// @placeholder
"foo": "Here is some placeholder text"
}
}The presence of @placeholder would tell Central not to send a message to Transifex. It'd also be easy to search the codebase for @placeholder.
Sometimes, we want to merge an initial version of a feature without deciding on the final text that users will see. This issue is to figure out a process for the
central-frontendrepo to facilitate that. The goals are:Here are a couple of ideas for how to implement this:
Idea 1. Just specify raw text in the
<template>, followed by a word likeTODO. We could search the codebase for TODO.Idea 2. Continue to use
$t()in the<template>, but make it possible to add messages to<i18n>blocks without sending them to Transifex. One way to do this would be via magic comments, which our Transifex scripts already know how to identify. Something like:The presence of
@placeholderwould tell Central not to send a message to Transifex. It'd also be easy to search the codebase for@placeholder.