Checks and fixups

The quality checks help catch common translator errors, ensuring the translation is in good shape. The checks can be ignored in case of false positives.

Once submitting a translation with a failing check, this is immediately shown to the user:

Automatic fixups

In addition to Quality checks, Weblate can fix some common errors in translated strings automatically. Use it with caution to not have it add errors.

See also

AUTOFIX_LIST

Trailing ellipsis replacer

Replace trailing dots (...) with an ellipsis (…) to make it consistent with the source string.

Zero-width space removal

Zero width space is typically not desired in the translation. This fix will remove it unless it is present in the source string as well.

Control characters removal

Removes any control characters from the translation.

Devanagari danda

Replaces wrong full stop in Devanagari by Devanagari danda (।).

Punctuation spacing

New in version 5.3.

Ensures French and Breton use correct punctuation spacing.

This fixup can be disable via ignore-punctuation-spacing flag (which also disables Punctuation spacing).

Unsafe HTML cleanup

When turned on using a safe-html flag it sanitizes HTML markup.

See also

Unsafe HTML

Trailing and leading whitespace fixer

Makes leading and trailing whitespace consistent with the source string. The behavior can be fine-tuned using ignore-begin-space and ignore-end-space flags to skip processing parts of the string.

Quality checks

Weblate employs a wide range of quality checks on strings. The following section describes them all in further detail. There are also language specific checks. Please file a bug if anything is reported in error.

See also

CHECK_LIST, Customizing behavior using flags

Translation checks

Executed upon every translation change, helping translators maintain good quality translations.

BBCode markup

Summary:

BBCode in translation does not match source

Scope:

translated strings

Check class:

weblate.checks.markup.BBCodeCheck

Check identifier:

bbcode

Flag to ignore:

ignore-bbcode

BBCode represents simple markup, like for example highlighting important parts of a message in bold font, or italics.

This check ensures they are also found in translation.

Note

The method for detecting BBCode is currently quite simple so this check might produce false positives.

Consecutive duplicated words

New in version 4.1.

Summary:

Text contains the same word twice in a row:

Scope:

translated strings

Check class:

weblate.checks.duplicate.DuplicateCheck

Check identifier:

duplicate

Flag to ignore:

ignore-duplicate

Checks that no consecutive duplicate words occur in a translation. This usually indicates a mistake in the translation.

Hint

This check includes language specific rules to avoid false positives. In case it triggers falsely in your case, let us know. See Reporting issues in Weblate.

Does not follow glossary

New in version 4.5.

Summary:

The translation does not follow terms defined in a glossary.

Scope:

translated strings

Check class:

weblate.checks.glossary.GlossaryCheck

Check identifier:

check_glossary

Flag to enable:

check-glossary

Flag to ignore:

ignore-check-glossary

This check has to be turned on using check-glossary flag (see Customizing behavior using flags). Please consider following prior to enabling it:

• It does exact string matching, the glossary is expected to contain terms in all variants.

• Checking each string against glossary is expensive, it will slow down any operation in Weblate which involves running checks like importing strings or translating.

• It also utilizes untranslatable glossary terms in Unchanged translation.

See also

Glossary, Customizing behavior using flags, Translation flags

Double space

Summary:

Translation contains double space

Scope:

translated strings

Check class:

weblate.checks.chars.DoubleSpaceCheck

Check identifier:

double_space

Flag to ignore:

ignore-double-space

Checks that double space is present in translation to avoid false positives on other space-related checks.

Check is false when double space is found in source meaning double space is intentional.

Fluent parts

New in version 5.0.

Summary:

Fluent parts should match

Scope:

translated strings

Check class:

weblate.checks.fluent.parts.FluentPartsCheck

Check identifier:

fluent-parts

Flag to enable:

fluent-parts

Flag to ignore:

ignore-fluent-parts

Each Fluent Message can have an optional value (the main text content), and optional attributes, each of which is a “part” of the Message. In Weblate, all these parts appear within the same block, using Fluent-like syntax to specify the attributes. For example:

This is the Message value
.title = This is the title attribute
.alt = This is the alt attribute

This check ensures that the translated Message also has a value if the source Message has one, or no value if the source has none. This also checks that the same attributes used in the source Message also appear in the translation, with no additions.

Note

This check is not applied to Fluent Terms since Terms always have a value, and Term attributes tend to be locale-specific (used for grammar rules, etc.), and are not expected to appear in all translations.

See also

Fluent Attributes

Fluent references

New in version 5.0.

Summary:

Fluent references should match

Scope:

translated strings

Check class:

weblate.checks.fluent.references.FluentReferencesCheck

Check identifier:

fluent-references

Flag to enable:

fluent-references

Flag to ignore:

ignore-fluent-references

A Fluent Message or Term can reference another Message, Term, Attribute, or a variable. For example:

Here is a { message }, a { message.attribute } a { -term } and a { $variable }.
Within a function { NUMBER($num, minimumFractionDigits: 2) }

Generally, translated Messages or Terms are expected to contain the same references as the source, although not necessarily in the same order of appearance. So this check ensures that translations use the same references in their value as the source value, the same number of times, and with no additions. For Messages, this will also check that each Attribute in the translation uses the same references as the matching Attribute in the source.

When the source or translation contains Fluent Select Expressions, then each possible variant in the source must be matched with at least one variant in the translation with the same references, and vice versa.

Moreover, if a variable reference appears both in the Select Expression’s selector and within one of its variants, then all variants may also be considered as if they also contain that reference. The assumption being that the variant’s key may have made the reference redundant for that variant. For example:

{ $num ->
    [one] an apple
   *[other] { $num } apples
}

Here, for the purposes of this check, the [one] variant will also be considered to contain the $num reference.

However, a reference within the Select Expression’s selector, which can only be a variable of a Term Attribute in Fluent’s syntax, will not by itself count as a required reference because they do not form the actual text content of the string that the end-user will see, and the presence of a Select Expression is considered locale-specific. For example:

{ -term.starts-with-vowel ->
    [yes] an { -term }
   *[no] a { -term }
}

Here a reference to -term.starts-with-vowel is not expected to appear in translations, but a reference to -term is.

See also

Fluent Variables Fluent Message and Term references Fluent Select Expressions

Fluent translation inner HTML

New in version 5.0.

Summary:

Fluent target should be valid inner HTML that matches

Scope:

translated strings

Check class:

weblate.checks.fluent.inner_html.FluentTargetInnerHTMLCheck

Check identifier:

fluent-target-inner-html

Flag to enable:

fluent-target-inner-html

Flag to ignore:

ignore-fluent-target-inner-html

This check will verify that the translated value of a Message or Term contains the same HTML elements as the source value.

First, if the source value fails the Fluent source inner HTML check, then this check will do nothing. Otherwise, the translated value will also be checked under the same conditions.

Second, the HTML elements found in the translated value will be compared against the HTML elements found in the source value. Two elements will match if they share the exact same tag name, the exact same attributes and values, and all their ancestors match in the same way. This check will ensure that all the elements in the source appear somewhere in the translation, with the same number of appearances, and with no additional elements added. When there are multiple elements in the value, they need not appear in the same order in the translation value.

When the source or translation contains Fluent Select Expressions, then each possible variant in the source must be matched with at least one variant in the translation with the same HTML elements, and vice versa.

When using Fluent in combination with the Fluent DOM package, this check will ensure that the translation also includes any required data-l10n-name elements that appear in the source, or any of the allowed inline elements like <br>.

For example, the following source:

Source message <img data-l10n-name="icon"/> with icon

would match with:

Translated message <img data-l10n-name="icon"/> with icon

but not:

Translated message <img data-l10n-name="new-val"/> with icon

nor

Translated message <br data-l10n-name="icon"/> with no icon

See also

Fluent source inner HTML, Fluent DOM

Fluent translation syntax

New in version 5.0.

Summary:

Fluent syntax error in translation

Scope:

translated strings

Check class:

weblate.checks.fluent.syntax.FluentTargetSyntaxCheck

Check identifier:

fluent-target-syntax

Flag to enable:

fluent-target-syntax

Flag to ignore:

ignore-fluent-target-syntax

In Weblate, Fluent strings use Fluent syntax for references and variables, but also for more complex features like defining attributes and selector variants, including plurals. This check ensures that the syntax used in the translation will be valid for Fluent.

See also

Fluent source syntax, Fluent Syntax Guide Mozilla Basic Syntax Guide

Formatted strings

Checks that the formatting in strings is replicated between both source and translation. Omitting format strings in translation usually causes severe problems, so the formatting in strings should usually match the source.

Weblate supports checking format strings in several languages. The check is not enabled automatically, only if a string is flagged appropriately (e.g. c-format for C format). Gettext adds this automatically, but you will probably have to add it manually for other file formats or if your PO files are not generated by xgettext.

Most of the format checks allow omitting format strings for plural forms having a single count. This allows translators to write nicer strings for these cases (One apple instead of %d apple). Turn this off by adding strict-format flag.

The flags can be customized per string (see Additional info on source strings) or in a Component configuration. Having it defined per component is simpler, but it can lead to false positives in case the string is not interpreted as a formatting string, but format string syntax happens to be used.

Hint

In case specific format check is not available in Weblate, you can use generic Placeholders.

Besides checking, this will also highlight the formatting strings to easily insert them into translated strings:

AngularJS interpolation string

Summary:

AngularJS interpolation strings do not match source

Scope:

translated strings

Check class:

weblate.checks.angularjs.AngularJSInterpolationCheck

Check identifier:

angularjs_format

Flag to enable:

angularjs-format

Flag to ignore:

ignore-angularjs-format

Named format string example:

Your balance is {{amount}} {{ currency }}

See also

Formatted strings, AngularJS text interpolation

C format

Summary:

C format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.CFormatCheck

Check identifier:

c_format

Flag to enable:

c-format

Flag to ignore:

ignore-c-format

Simple format string example:

There are %d apples

Position format string example:

Your balance is %1$d %2$s

See also

Formatted strings,

C format strings, C printf format

C# format

Summary:

C# format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.CSharpFormatCheck

Check identifier:

c_sharp_format

Flag to enable:

c-sharp-format

Flag to ignore:

ignore-c-sharp-format

Position format string example:

There are {0} apples

See also

Formatted strings, C# String Format

ECMAScript template literals

Summary:

ECMAScript template literals do not match source

Scope:

translated strings

Check class:

weblate.checks.format.ESTemplateLiteralsCheck

Check identifier:

es_format

Flag to enable:

es-format

Flag to ignore:

ignore-es-format

Interpolation example:

There are ${number} apples

See also

Formatted strings, Template literals

i18next interpolation

New in version 4.0.

Summary:

The i18next interpolation does not match source

Scope:

translated strings

Check class:

weblate.checks.format.I18NextInterpolationCheck

Check identifier:

i18next_interpolation

Flag to enable:

i18next-interpolation

Flag to ignore:

ignore-i18next-interpolation

Interpolation example:

There are {{number}} apples

Nesting example:

There are $t(number) apples

See also

Formatted strings, i18next interpolation

ICU MessageFormat

New in version 4.9.

Summary:

Syntax errors and/or placeholder mismatches in ICU MessageFormat strings.

Scope:

translated strings

Check class:

weblate.checks.icu.ICUMessageFormatCheck

Check identifier:

icu_message_format

Flag to enable:

icu-message-format

Flag to ignore:

ignore-icu-message-format

Interpolation example:

There {number, plural, one {is one apple} other {are # apples}}.

This check has support for both pure ICU MessageFormat messages as well as ICU with simple XML tags. You can configure the behavior of this check by using icu-flags:*, either by opting into XML support or by disabling certain sub-checks. For example, the following flag enables XML support while disabling validation of plural sub-messages:

icu-message-format, icu-flags:xml:-plural_selectors

xml	Enable support for simple XML tags. By default, XML tags are parsed loosely. Stray < characters are ignored if they are not reasonably part of a tag.
strict-xml	Enable support for strict XML tags. All < characters must be escaped if they are not part of a tag.
-highlight	Disable highlighting placeholders in the editor.
-require_other	Disable requiring sub-messages to have an other selector.
-submessage_selectors	Skip checking that sub-message selectors match the source.
-types	Skip checking that placeholder types match the source.
-extra	Skip checking that no placeholders are present that were not present in the source string.
-missing	Skip checking that no placeholders are missing that were present in the source string.

Additionally, when strict-xml is not enabled but xml is enabled, you can use the icu-tag-prefix:PREFIX flag to require that all XML tags start with a specific string. For example, the following flag will only allow XML tags to be matched if they start with <x::

icu-message-format, icu-flags:xml, icu-tag-prefix:"x:"

This would match <x:link>click here</x:link> but not <strong>this</strong>.

See also

ICU MessageFormat syntax, Formatted strings, ICU: Formatting Messages, Format.JS: Message Syntax

Java format

Summary:

Java format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.JavaFormatCheck

Check identifier:

java_printf_format

Flag to enable:

java-printf-format

Flag to ignore:

ignore-java-printf-format

Simple format string example:

There are %d apples

Position format string example:

Your balance is %1$d %2$s

Changed in version 4.14: This used to be toggled by java-format flag, it was changed for consistency with GNU gettext.

See also

Formatted strings, Java Format Strings

Java MessageFormat

Summary:

Java MessageFormat string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.JavaMessageFormatCheck

Check identifier:

java_format

Flag to enable unconditionally:

java-format

Flag to enable autodetection:

auto-java-messageformat enables check only if there is a format string in the source

Flag to ignore:

ignore-java-format

Position format string example:

There are {0} apples

Changed in version 4.14: This used to be toggled by java-messageformat flag, it was changed for consistency with GNU gettext.

This check validates that format string is valid for the Java MessageFormat class. Besides matching format strings in the curly braces, it also verifies single quotes as they have a special meaning. Whenever writing single quote, it should be written as ''. When not paired, it is treated as beginning of quoting and will not be shown when rendering the string.

See also

Formatted strings, Java MessageFormat

JavaScript format

Summary:

JavaScript format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.JavaScriptFormatCheck

Check identifier:

javascript_format

Flag to enable:

javascript-format

Flag to ignore:

ignore-javascript-format

Simple format string example:

There are %d apples

See also

Formatted strings, JavaScript formatting strings

Lua format

Summary:

Lua format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.LuaFormatCheck

Check identifier:

lua_format

Flag to enable:

lua-format

Flag to ignore:

ignore-lua-format

Simple format string example:

There are %d apples

See also

Formatted strings, Lua formatting strings

Object Pascal format

Summary:

Object Pascal format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.ObjectPascalFormatCheck

Check identifier:

object_pascal_format

Flag to enable:

object-pascal-format

Flag to ignore:

ignore-object-pascal-format

Simple format string example:

There are %d apples

See also

Formatted strings, Object Pascal formatting strings, Free Pascal formatting strings Delphi formatting strings

Percent placeholders

New in version 4.0.

Summary:

The percent placeholders do not match source

Scope:

translated strings

Check class:

weblate.checks.format.PercentPlaceholdersCheck

Check identifier:

percent_placeholders

Flag to enable:

percent-placeholders

Flag to ignore:

ignore-percent-placeholders

Simple format string example:

There are %number% apples

See also

Formatted strings,

Perl format

Summary:

Perl format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.PerlFormatCheck

Check identifier:

perl_format

Flag to enable:

perl-format

Flag to ignore:

ignore-perl-format

Simple format string example:

There are %d apples

Position format string example:

Your balance is %1$d %2$s

See also

Formatted strings, Perl sprintf, Perl Format Strings

PHP format

Summary:

PHP format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.PHPFormatCheck

Check identifier:

php_format

Flag to enable:

php-format

Flag to ignore:

ignore-php-format

Simple format string example:

There are %d apples

Position format string example:

Your balance is %1$d %2$s

See also

Formatted strings, PHP sprintf documentation, PHP Format Strings

Python brace format

Summary:

Python brace format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.PythonBraceFormatCheck

Check identifier:

python_brace_format

Flag to enable:

python-brace-format

Flag to ignore:

ignore-python-brace-format

Simple format string:

There are {} apples

Named format string example:

Your balance is {amount} {currency}

See also

Formatted strings, Python brace format, Python Format Strings

Python format

Summary:

Python format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.PythonFormatCheck

Check identifier:

python_format

Flag to enable:

python-format

Flag to ignore:

ignore-python-format

Simple format string:

There are %d apples

Named format string example:

Your balance is %(amount)d %(currency)s

See also

Formatted strings, Python string formatting, Python Format Strings

Qt format

Summary:

Qt format string does not match source

Scope:

translated strings

Check class:

weblate.checks.qt.QtFormatCheck

Check identifier:

qt_format

Flag to enable:

qt-format

Flag to ignore:

ignore-qt-format

Position format string example:

There are %1 apples

See also

Formatted strings, Qt QString::arg()

Qt plural format

Summary:

Qt plural format string does not match source

Scope:

translated strings

Check class:

weblate.checks.qt.QtPluralCheck

Check identifier:

qt_plural_format

Flag to enable:

qt-plural-format

Flag to ignore:

ignore-qt-plural-format

Plural format string example:

There are %Ln apple(s)

See also

Formatted strings, Qt i18n guide

Ruby format

Summary:

Ruby format string does not match source

Scope:

translated strings

Check class:

weblate.checks.ruby.RubyFormatCheck

Check identifier:

ruby_format

Flag to enable:

ruby-format

Flag to ignore:

ignore-ruby-format

Simple format string example:

There are %d apples

Position format string example:

Your balance is %1$f %2$s

Named format string example:

Your balance is %+.2<amount>f %<currency>s

Named template string:

Your balance is %{amount} %{currency}

See also

Formatted strings, Ruby Kernel#sprintf

Scheme format

Summary:

Scheme format string does not match source

Scope:

translated strings

Check class:

weblate.checks.format.SchemeFormatCheck

Check identifier:

scheme_format

Flag to enable:

scheme-format

Flag to ignore:

ignore-scheme-format

Simple format string example:

There are ~d apples

See also

Formatted strings, Srfi 28, Chicken Scheme format, Guile Scheme formatted output

Vue I18n formatting

Summary:

The Vue I18n formatting does not match source

Scope:

translated strings

Check class:

weblate.checks.format.VueFormattingCheck

Check identifier:

vue_format

Flag to enable:

vue-format

Flag to ignore:

ignore-vue-format

Named formatting:

There are {count} apples

Rails i18n formatting:

There are %{count} apples

Linked locale messages:

@:message.dio @:message.the_world!

See also

Formatted strings, Vue I18n Formatting, Vue I18n Linked locale messages

Has been translated

Summary:

This string has been translated in the past

Scope:

all strings

Check class:

weblate.checks.consistency.TranslatedCheck

Check identifier:

translated

Flag to ignore:

ignore-translated

Means a string has been translated already. This can happen when the translations have been reverted in VCS or lost otherwise.

Inconsistent

Summary:

This string has more than one translation in this project or is untranslated in some components.

Scope:

all strings

Check class:

weblate.checks.consistency.ConsistencyCheck

Check identifier:

inconsistent

Flag to ignore:

ignore-inconsistent

Weblate checks translations of the same string across all translation within a project to help you keep consistent translations.

The check fails on differing translations of one string within a project. This can also lead to inconsistencies in displayed checks. You can find other translations of this string on the Other occurrences tab.

This check applies to all components in a project that have Allow translation propagation turned on.

Hint

For performance reasons, the check might not find all inconsistencies, it limits number of matches.

Note

This check also fires in case the string is translated in one component and not in another. It can be used as a quick way to manually handle strings which are untranslated in some components just by clicking on the Use this translation button displayed on each line in the Other occurrences tab.

You can use Automatic translation add-on to automate translating of newly added strings which are already translated in another component.

See also

Keeping translations same across components

Kashida letter used

Summary:

The decorative kashida letters should not be used

Scope:

translated strings

Check class:

weblate.checks.chars.KashidaCheck

Check identifier:

kashida

Flag to ignore:

ignore-kashida

The decorative Kashida letters should not be used in translation. These are also known as Tatweel.

See also

Kashida on Wikipedia

Markdown links

Summary:

Markdown links do not match source

Scope:

translated strings

Check class:

weblate.checks.markup.MarkdownLinkCheck

Check identifier:

md-link

Flag to enable:

md-text

Flag to ignore:

ignore-md-link

Markdown links do not match source.

See also

Markdown links

Markdown references

Summary:

Markdown link references do not match source

Scope:

translated strings

Check class:

weblate.checks.markup.MarkdownRefLinkCheck

Check identifier:

md-reflink

Flag to enable:

md-text

Flag to ignore:

ignore-md-reflink

Markdown link references do not match source.

See also

Markdown links

Markdown syntax

Summary:

Markdown syntax does not match source

Scope:

translated strings

Check class:

weblate.checks.markup.MarkdownSyntaxCheck

Check identifier:

md-syntax

Flag to enable:

md-text

Flag to ignore:

ignore-md-syntax

Markdown syntax does not match source

See also

Markdown span elements

Maximum length of translation

Summary:

Translation should not exceed given length

Scope:

translated strings

Check class:

weblate.checks.chars.MaxLengthCheck

Check identifier:

max-length

Flag to enable:

max-length

Flag to ignore:

ignore-max-length

Checks that translations are of acceptable length to fit available space. This only checks for the length of translation characters.

Unlike the other checks, the flag should be set as a key:value pair like max-length:100.

Hint

This check looks at number of chars, what might not be the best metric when using proportional fonts to render the text. The Maximum size of translation check does check actual rendering of the text.

The replacements: flag might be also useful to expand placeables before checking the string.

When xml-text flag is also used, the length calculation ignores XML tags.

Maximum size of translation

Summary:

Translation rendered text should not exceed given size

Scope:

translated strings

Check class:

weblate.checks.render.MaxSizeCheck

Check identifier:

max-size

Flag to enable:

max-size

Flag to ignore:

ignore-max-size

Translation rendered text should not exceed given size. It renders the text with line wrapping and checks if it fits into given boundaries.

This check needs one or two parameters - maximal width and maximal number of lines. In case the number of lines is not provided, one line text is considered.

You can also configure used font by font-* directives (see Customizing behavior using flags), for example following translation flags say that the text rendered with ubuntu font size 22 should fit into two lines and 500 pixels:

max-size:500:2, font-family:ubuntu, font-size:22

Hint

You might want to set font-* directives in Component configuration to have the same font configured for all strings within a component. You can override those values per string in case you need to customize it per string.

The replacements: flag might be also useful to expand placeables before checking the string.

When xml-text flag is also used, the length calculation ignores XML tags.

See also

Managing fonts, Customizing behavior using flags, Maximum length of translation

Mismatched \n

Summary:

Number of \n literals in translation does not match source

Scope:

translated strings

Check class:

weblate.checks.chars.EscapedNewlineCountingCheck

Check identifier:

escaped_newline

Flag to ignore:

ignore-escaped-newline

Usually escaped newlines are important for formatting program output. Check fails if the number of \n literals in translation does not match the source.

Mismatched colon

Summary:

Source and translation do not both end with a colon

Scope:

translated strings

Check class:

weblate.checks.chars.EndColonCheck

Check identifier:

end_colon

Flag to ignore:

ignore-end-colon

Checks that colons are replicated between both source and translation. The presence of colons is also checked for various languages where they do not belong (Chinese or Japanese).

See also

Colon on Wikipedia

Mismatched ellipsis

Summary:

Source and translation do not both end with an ellipsis

Scope:

translated strings

Check class:

weblate.checks.chars.EndEllipsisCheck

Check identifier:

end_ellipsis

Flag to ignore:

ignore-end-ellipsis

Checks that trailing ellipses are replicated between both source and translation. This only checks for real ellipsis (…) not for three dots (...).

An ellipsis is usually rendered nicer than three dots in print, and sounds better with text-to-speech.

See also

Ellipsis on Wikipedia

Mismatched exclamation mark

Summary:

Source and translation do not both end with an exclamation mark

Scope:

translated strings

Check class:

weblate.checks.chars.EndExclamationCheck

Check identifier:

end_exclamation

Flag to ignore:

ignore-end-exclamation

Checks that exclamations are replicated between both source and translation. The presence of exclamation marks is also checked for various languages where they do not belong (Chinese, Japanese, Korean, Armenian, Limbu, Myanmar or Nko).

See also

Exclamation mark on Wikipedia

Mismatched full stop

Summary:

Source and translation do not both end with a full stop

Scope:

translated strings

Check class:

weblate.checks.chars.EndStopCheck

Check identifier:

end_stop

Flag to ignore:

ignore-end-stop

Checks that full stops are replicated between both source and translation. The presence of full stops is checked for various languages where they do not belong (Chinese, Japanese, Devanagari or Urdu).

See also

Full stop on Wikipedia

Mismatched question mark

Summary:

Source and translation do not both end with a question mark

Scope:

translated strings

Check class:

weblate.checks.chars.EndQuestionCheck

Check identifier:

end_question

Flag to ignore:

ignore-end-question

Checks that question marks are replicated between both source and translation. The presence of question marks is also checked for various languages where they do not belong (Armenian, Arabic, Chinese, Korean, Japanese, Ethiopic, Vai or Coptic).

See also

Question mark on Wikipedia

Mismatched semicolon

Summary:

Source and translation do not both end with a semicolon

Scope:

translated strings

Check class:

weblate.checks.chars.EndSemicolonCheck

Check identifier:

end_semicolon

Flag to ignore:

ignore-end-semicolon

Checks that semicolons at the end of sentences are replicated between both source and translation.

See also

Semicolon on Wikipedia

Mismatching line breaks

Summary:

Number of new lines in translation does not match source

Scope:

translated strings

Check class:

weblate.checks.chars.NewLineCountCheck

Check identifier:

newline-count

Flag to ignore:

ignore-newline-count

Usually newlines are important for formatting program output. Check fails if the number of new lines in translation does not match the source.

Missing plurals

Summary:

Some plural forms are untranslated

Scope:

translated strings

Check class:

weblate.checks.consistency.PluralsCheck

Check identifier:

plurals

Flag to ignore:

ignore-plurals

Checks that all plural forms of a source string have been translated. Specifics on how each plural form is used can be found in the string definition.

Failing to fill in plural forms will in some cases lead to displaying nothing when the plural form is in use.

Placeholders

Summary:

Translation is missing some placeholders

Scope:

translated strings

Check class:

weblate.checks.placeholders.PlaceholderCheck

Check identifier:

placeholders

Flag to enable:

placeholders

Flag to ignore:

ignore-placeholders

Changed in version 4.3: You can use regular expression as placeholder.

Changed in version 4.13: With the case-insensitive flag, the placeholders are not case-sensitive.

Translation is missing some placeholders. These are either extracted from the translation file or defined manually using placeholders flag, more can be separated with colon, strings with space can be quoted:

placeholders:$URL$:$TARGET$:"some long text"

In case you have some syntax for placeholders, you can use a regular expression:

placeholders:r"%[^% ]%"

You can also have case insensitive placeholders:

placeholders:$URL$:$TARGET$,case-insensitive

See also

Customizing behavior using flags

Punctuation spacing

Summary:

Missing non breakable space before double punctuation sign

Scope:

translated strings

Check class:

weblate.checks.chars.PunctuationSpacingCheck

Check identifier:

punctuation_spacing

Flag to ignore:

ignore-punctuation-spacing

Checks that there is non breakable space before double punctuation sign (exclamation mark, question mark, semicolon and colon). This rule is used only in a few selected languages like French or Breton, where space before double punctuation sign is a typographic rule.

See also

French and English spacing on Wikipedia

Regular expression

Summary:

Translation does not match regular expression

Scope:

translated strings

Check class:

weblate.checks.placeholders.RegexCheck

Check identifier:

regex

Flag to enable:

regex

Flag to ignore:

ignore-regex

Translation does not match regular expression. The expression is either extracted from the translation file or defined manually using regex flag:

regex:^foo|bar$

Reused translation

New in version 4.18.

Summary:

Different strings are translated the same.

Scope:

translated strings

Check class:

weblate.checks.consistency.ReusedCheck

Check identifier:

reused

Flag to ignore:

ignore-reused

Check that fails if the same translation is used on different source strings. Such translations can be intentional, but can also confuse users.

Same plurals

Summary:

Some plural forms are translated in the same way

Scope:

translated strings

Check class:

weblate.checks.consistency.SamePluralsCheck

Check identifier:

same-plurals

Flag to ignore:

ignore-same-plurals

Check that fails if some plural forms are duplicated in the translation. In most languages they have to be different.

Starting newline

Summary:

Source and translation do not both start with a newline

Scope:

translated strings

Check class:

weblate.checks.chars.BeginNewlineCheck

Check identifier:

begin_newline

Flag to ignore:

ignore-begin-newline

Newlines usually appear in source strings for good reason, omissions or additions can lead to formatting problems when the translated text is put to use.

See also

Trailing newline

Starting spaces

Summary:

Source and translation do not both start with same number of spaces

Scope:

translated strings

Check class:

weblate.checks.chars.BeginSpaceCheck

Check identifier:

begin_space

Flag to ignore:

ignore-begin-space

A space in the beginning of a string is usually used for indentation in the interface and thus important to keep.

Trailing newline

Summary:

Source and translation do not both end with a newline

Scope:

translated strings

Check class:

weblate.checks.chars.EndNewlineCheck

Check identifier:

end_newline

Flag to ignore:

ignore-end-newline

Newlines usually appear in source strings for good reason, omissions or additions can lead to formatting problems when the translated text is put to use.

See also

Starting newline

Trailing space

Summary:

Source and translation do not both end with a space

Scope:

translated strings

Check class:

weblate.checks.chars.EndSpaceCheck

Check identifier:

end_space

Flag to ignore:

ignore-end-space

Checks that trailing spaces are replicated between both source and translation.

Trailing space is usually utilized to space out neighbouring elements, so removing it might break layout.

Unchanged translation

Summary:

Source and translation are identical

Scope:

translated strings

Check class:

weblate.checks.same.SameCheck

Check identifier:

same

Flag to ignore:

ignore-same

Happens if the source and corresponding translation strings is identical, down to at least one of the plural forms. Some strings commonly found across all languages are ignored, and various markup is stripped. This reduces the number of false positives.

This check can help find strings mistakenly untranslated.

The default behavior of this check is to exclude words from the built-in blacklist from the checking. These are words which are frequently not being translated. This is useful to avoid false positives on short strings, which consist only of single word which is same in several languages. This blacklist can be disabled by adding strict-same flag to string or component.

Changed in version 4.17: With check-glossary flag (see Does not follow glossary), the untranslatable glossary terms are excluded from the checking.

See also

Does not follow glossary, Component configuration, Customizing behavior using flags

Unsafe HTML

Summary:

The translation uses unsafe HTML markup

Scope:

translated strings

Check class:

weblate.checks.markup.SafeHTMLCheck

Check identifier:

safe-html

Flag to enable:

safe-html

Flag to ignore:

ignore-safe-html

The translation uses unsafe HTML markup. This check has to be enabled using safe-html flag (see Customizing behavior using flags). There is also accompanied autofixer which can automatically sanitize the markup.

Hint

When md-text flag is also used, the Markdown style links are also allowed.

See also

The HTML check is performed by the Ammonia library.

URL

Summary:

The translation does not contain an URL

Scope:

translated strings

Check class:

weblate.checks.markup.URLCheck

Check identifier:

url

Flag to enable:

url

Flag to ignore:

ignore-url

The translation does not contain an URL. This is triggered only in case the unit is marked as containing URL. In that case the translation has to be a valid URL.

XML markup

Summary:

XML tags in translation do not match source

Scope:

translated strings

Check class:

weblate.checks.markup.XMLTagsCheck

Check identifier:

xml-tags

Flag to ignore:

ignore-xml-tags

This usually means the resulting output will look different. In most cases this is not a desired result from changing the translation, but occasionally it is.

Checks that XML tags are replicated between both source and translation.

The check is automatically enabled for XML like strings. You might need to add xml-text flag in some cases to force turning it on.

Note

This check is disabled by the safe-html flag as the HTML cleanup done by it can produce HTML markup which is not valid XML.

XML syntax

Summary:

The translation is not valid XML

Scope:

translated strings

Check class:

weblate.checks.markup.XMLValidityCheck

Check identifier:

xml-invalid

Flag to ignore:

ignore-xml-invalid

The XML markup is not valid.

The check is automatically enabled for XML like strings. You might need to add xml-text flag in some cases to force turning it on.

Note

This check is disabled by the safe-html flag as the HTML cleanup done by it can produce HTML markup which is not valid XML.

Zero-width space

Summary:

Translation contains extra zero-width space character

Scope:

translated strings

Check class:

weblate.checks.chars.ZeroWidthSpaceCheck

Check identifier:

zero-width-space

Flag to ignore:

ignore-zero-width-space

Zero-width space (<U+200B>) characters are used to break messages within words (word wrapping).

As they are usually inserted by mistake, this check is triggered once they are present in translation. Some programs might have problems when this character is used.

See also

Zero width space on Wikipedia

Source checks

Source checks can help developers improve the quality of source strings.

Ellipsis

Summary:

The string uses three dots (…) instead of an ellipsis character (…)

Scope:

source strings

Check class:

weblate.checks.source.EllipsisCheck

Check identifier:

ellipsis

Flag to ignore:

ignore-ellipsis

This fails when the string uses three dots (...) when it should use an ellipsis character (…).

Using the Unicode character is in most cases the better approach and looks better rendered, and may sound better with text-to-speech.

See also

Ellipsis on Wikipedia

Fluent source inner HTML

New in version 5.0.

Summary:

Fluent source should be valid inner HTML

Scope:

source strings

Check class:

weblate.checks.fluent.inner_html.FluentSourceInnerHTMLCheck

Check identifier:

fluent-source-inner-html

Flag to enable:

fluent-source-inner-html

Flag to ignore:

ignore-fluent-source-inner-html

Fluent is often used in contexts where the value for a Message (or Term) is meant to be used directly as .innerHTML (rather than .textContent) for some HTML element. For example, when using the Fluent DOM package.

The aim of this check is to predict how the value will be parsed as inner HTML, assuming a HTML5 conforming parser, to catch cases where there would be some “unintended” loss of the string, without being too strict about technical parsing errors that do not lead to a loss of the string.

This check is applied to the value of Fluent Messages or Terms, but not their Attributes. For Messages, the Fluent Attributes are often just HTML attribute values, so can be arbitrary strings. For Terms, the Fluent Attributes are often language properties that can only be referenced in the selectors of Fluent Select Expressions.

Generally, most Fluent values are not expected to contain any HTML markup. Therefore, this check does not expect or want translators and developers to have to care about strictly avoiding any technical HTML5 parsing errors (let alone XHTML parsing errors). Instead, this check will just want to warn them when they may have unintentionally opened a HTML tag or inserted a character reference.

Moreover, for the Fluent values that intentionally contain HTML tags or character references, this check will verify some “good practices”, such as matching closing and ending tags, valid character references, and quoted attribute values. In addition, whilst the HTML5 specification technically allows for quite arbitrary tag and attribute names, this check will restrain them to some basic ASCII values that should cover the standard HTML5 element tags and attributes, as well as allow some custom element or attribute names. This is partially to ensure that the user is using HTML intentionally.

Examples:

Value	Warns?	Reason
three<four	yes	The <four part would be lost as .innerHTML.
three < four	no	The .innerHTML would match the .textContent.
three <four>	yes	Missing a closing tag.
three <four/>	yes	four is not a HTML void element, so should not self-close.
<a-b>text</a-b>	no	Custom element tag with a matching closing tag.
a <img/> b	no	img is a HTML void element. Self-closing is allowed.
a <br> b	no	br is a HTML void element.
<img class=a/>	yes	The attribute value is not quoted.
<aØ attr=''/>	yes	Non-ASCII tag name.
kind&ethical	yes	The &eth part would be converted to ð.
kind&eth;ical	no	The character reference seems to be intentional.
three&lte;four	yes	The &lte; part would be converted to <e;.
three&lf;four	yes	The character reference is not valid.
three<{ $val }	yes	The Fluent variable may unintentionally become a tag.
&l{ $val }	yes	The Fluent variable may unintentionally become a character reference.

Note

This check will not ensure the inner HTML is safe or sanitized, and is not meant to protect against malicious attempts to alter the inner HTML. Moreover, it should be remembered that Fluent variables and references may expand to arbitrary strings, so could expand to arbitrary HTML unless they are escaped. As an exception, a < or & character before a Fluent reference will trigger this check since even an escaped value could lead to unexpected results.

Note

The Fluent DOM package has further limitations, such as allowed tags and attributes, which this check will not enforce.

See also

Fluent translation inner HTML, Fluent DOM

Fluent source syntax

New in version 5.0.

Summary:

Fluent syntax error in source

Scope:

source strings

Check class:

weblate.checks.fluent.syntax.FluentSourceSyntaxCheck

Check identifier:

fluent-source-syntax

Flag to enable:

fluent-source-syntax

Flag to ignore:

ignore-fluent-source-syntax

In Weblate, Fluent strings use Fluent syntax for references and variables, but also for more complex features like defining attributes and selector variants, including plurals. This check ensures that the syntax used in source will be valid for Fluent.

See also

Fluent translation syntax, Fluent Syntax Guide Mozilla Basic Syntax Guide

ICU MessageFormat syntax

New in version 4.9.

Summary:

Syntax errors in ICU MessageFormat strings.

Scope:

source strings

Check class:

weblate.checks.icu.ICUSourceCheck

Check identifier:

icu_message_format_syntax

Flag to enable:

icu-message-format

Flag to ignore:

ignore-icu-message-format

See also

ICU MessageFormat

Long untranslated

New in version 4.1.

Summary:

The string has not been translated for a long time

Scope:

source strings

Check class:

weblate.checks.source.LongUntranslatedCheck

Check identifier:

long_untranslated

Flag to ignore:

ignore-long-untranslated

When the string has not been translated for a long time, it can indicate a problem in a source string making it hard to translate.

Multiple failing checks

Summary:

The translations in several languages have failing checks

Scope:

source strings

Check class:

weblate.checks.source.MultipleFailingCheck

Check identifier:

multiple_failures

Flag to ignore:

ignore-multiple-failures

Numerous translations of this string have failing quality checks. This is usually an indication that something could be done to improve the source string.

This check failing can quite often be caused by a missing full stop at the end of a sentence, or similar minor issues which translators tend to fix in translation, while it would be better to fix it in the source string.

Multiple unnamed variables

New in version 4.1.

Summary:

There are multiple unnamed variables in the string, making it impossible for translators to reorder them

Scope:

source strings

Check class:

weblate.checks.format.MultipleUnnamedFormatsCheck

Check identifier:

unnamed_format

Flag to ignore:

ignore-unnamed-format

There are multiple unnamed variables in the string, making it impossible for translators to reorder them.

Consider using named variables instead to allow translators to reorder them.

Unpluralised

Summary:

The string is used as plural, but not using plural forms

Scope:

source strings

Check class:

weblate.checks.source.OptionalPluralCheck

Check identifier:

optional_plural

Flag to ignore:

ignore-optional-plural

The string is used as a plural, but does not use plural forms. In case your translation system supports this, you should use the plural aware variant of it.

For example with Gettext in Python it could be:

from gettext import ngettext

print(ngettext("Selected %d file", "Selected %d files", files) % files)