Home/Cloud Services/On-Premises

Import and Export to Word On-Premises

# 1.21.0 (2025-04-08)

# Import from Word

# Features

Added generic font families to fonts imported from MS Word

MS Word can include many fonts that are not available in the browser, but are part of the user’s OS.
To make sure that fonts rendered by the browser resemble what the user sees in Word, we have introduced generic
font families in outputted font stacks. For example, instead of setting font-family to Calibri, the converter
now sets it to Calibri, sans-serif.

# Bug fixes

Fixed an issue with incorrectly recognized multi-level lists

When a multi-level list definition included other list item types that were not used in the list itself,
the converter produced a normal list. This optimization makes sure that multi-level lists with unused other
list type definitions are still treated as multi-level.

# Export to Word

# Features

Added table of contents support

Export for Word now supports the table of contents feature that is converted into a proper MS Word feature.

# Bug fixes

Fixed missing support for Windows/DOS line endings

Added support for Windows/DOS line endings, i.e. \r\n.
Fixed failing conversion when converting unsupported drawings

In rare cases when an unsupported type of drawing was added to the document as a suggestion, the import suggestion failed.

# Common changes

Improved memory consumption

The converter now consumes less memory while processing large documents.
Updated node version to 22.14.0

The node version has been updated to 22.14.0 improving the security, reliability
and performance of the application.
Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.20.1 (2025-02-11)

# Import from Word

# Bug fixes

Fixed incorrect indentation in paragraphs with disabled numbering

When a paragraph inherited a numbering definition containing paragraph indentation from a style definition, but had that numbering explicitly disabled in direct formatting, the indentation was still being applied. In this release, this bug was fixed, and such paragraphs no longer include the indentation inherited from a numbering definition.
Fixed an exception thrown in documents containing references to non-existent abstract numbering definitions

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.20.0 (2024-12-17)

# Import from Word

# Features

Added multi-level lists

With this release, we introduced support for legal style numbering. That kind of multi-level list refers to a numbering format used primarily in legal documents, characterized by a hierarchical structure where each section, subsection, and clause is distinctly numbered (e.g., 1, 1.1, 1.1.1) to facilitate clear referencing and organization. This feature is fully integrated with CKEditor 5 Multi-level lists.
Added document lists

Paragraphs that follow list items with equal or greater indentation are now automatically included as part of the list item, preserving complex list structures from Word documents.

# Bug fixes

Fixed an issue with incorrectly numbered list items

Fixed Word list import to properly handle custom number overrides, ensuring that lists with manually adjusted numbering sequences are accurately preserved in the resulting HTML output.
Fixed HTML encoding for selected attributes

Some attributes could generate invalid HTML if not properly encoded. This bugfix ensures that the produced HTML is fully HTML 5 compliant.
Updated URL protocol validation

Updated protocol handling to omit specific non-web protocols that could produce non-compliant HTML.

# Other changes

Improved file validation in Import from Word

In case of incompatible file import a proper error message will be returned.

# Export to Word

# Features

Added multi-level lists

With this release, we introduced support for legal style numbering. That kind of multi-level list refers to a numbering format used primarily in legal documents, characterized by a hierarchical structure where each section, subsection, and clause is distinctly numbered (e.g., 1, 1.1, 1.1.1) to facilitate clear referencing and organization. This feature is fully integrated with CKEditor 5 Multi-level lists.
Added document lists

List items containing block elements are now automatically flattened when exporting to Word, with nested content converted into properly indented paragraphs following their respective list items.
Improved image spacing

Export to Word is now capable of recognizing image margins for float elements (CSS float property), maintaining proper spacing between floating images and surrounding text.
Added support for image merge fields

Added support for the merge fields within image source attributes, allowing dynamic image references when field values are provided. Note: due to Word limitations, this cannot generate new image fields in the document on its own, you have to provide merge field value via the merge_fields.data configuration option.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.19.0 (2024-10-01)

# Import form Word

# Features

Added support for detecting text content language

The text content language is now preserved by the converter and output as the appropriate lang attribute.
This feature is fully compatible with CKEditor 5’s Text part language feature.

# Bug fixes

Fixed incorrectly converted track changes newline insertions in lists

When the last item in a nested list included a newline insertion, it was incorrectly converted and became
attached to the parent list. This issue has been resolved.

# Export to Word

# Features

Added watermark support

You can now add a semi-transparent image watermark to each page of an exported Word document, ideal for branding, copyright protection, or decorative purposes. This feature is particularly useful for professional documents such as reports, contracts, or marketing materials. For more details, see the API watermark configuration option.
Added bookmarks support

The converter now transforms elements with the id attribute into Word document bookmarks, enabling quick navigation within the document.
Added support for detecting text content language

Text content language is now retained in exported Word documents, compatible with CKEditor 5’s Text part language feature.

# Bug fixes

Fixed table rendering across multiple pages

Fixed an issue where tables spanning multiple pages had incorrect break points if the CSS did not specify a parent element font size.
Table height is now calculated only when explicitly defined in CSS, allowing Word’s rendering engine to handle table pagination more accurately.
Fixed issue with empty documents not preserving document language

Empty documents now retain the specified content language in the exported Word document, ensuring a better editing
experience even when no initial content is present.
Fixed issue with images and links pointing to the same resource

Documents containing the same src attribute for images and href for links are now correctly generated without breaking
the document structure.
Fixed attribute suggestions breaking the conversion ( EtW v1 version )

The unsupported attribute suggestions are removed from the document.

# Other changes

Improved spacing between adjacent tables and block quotes

When HTML contains two adjacent tables or block quotes, we insert a paragraph in-between to prevent Word from merging these elements.
These paragraphs now use the “No Spacing” style, reducing formatting discrepancies.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.18.0 (2024-08-06)

# Import form Word

# Features

Added merge fields support

Alongside the CKEditor 5 merge fields release, we have introduced merge fields for import from Word. Merge fields found in Word documents are preserved and outputted, with an option to define merge field prefixes and suffixes.
You can find more information in the REST API documentation available under /v2/convert/docs#section/Import-from-Word/Configuration.
Added the timezone option to help convert dates of the collaboration features more accurately

To properly offset the date and time, the timezone option is used to inform the converter about the time zone in which the document was created. This option is required for the collaboration features to accurately calculate the created_at date.
You can find more information in the REST API documentation available under /v2/convert/docs#section/Import-from-Word/Configuration.
Collaboration features (comments and suggestions) will return the creation date.

Both comments and suggestions have been improved to include the annotation creation date as the created_at property.
Dates are ISO 8601 compatible.

# Bug fixes

Improved table border resolution

In some cases, imported table borders were incorrectly applied or missing.
Changed the japaneseDigitalTenThousand (Japanese Digital Ten Thousand Counting System) list style type to CSS cjk-decimal.

Originally, that style was converted to japanese-informal, which did not properly match this list style type.

# Export to Word

# Features

Added complete support for table borders formatting

Now a change of table border color, style, and width, as well as creating tables with hidden borders to layout the content are supported.
Added merge fields support

Alongside the CKEditor 5 merge fields release, we have introduced merge fields for export to Word. Exported merge fields integrate well with the Word mail merge feature. You can also replace merge fields on the fly by providing merge field data in JSON format.
You can find more information in the REST API documentation available under /v2/convert/docs#section/Export-to-Word/Configuration.
Added support for document content language

The document content language can be changed now by configuring the document.language option.
This accepts a valid BCP 47 language tag.
You can find more information in the REST API documentation available under /v2/convert/docs#section/Export-to-Word/Configuration.

# Bug fixes

Fixed an issue with images positioned incorrectly with text wrap

Images with wrap text or side image alignment in the document list were positioned incorrectly. After the fix, image positioning for these types of images is more precise.
Fixed image caption positioning.

A recent update in MS Word caused an issue where image captions would occupy the entire height of the image. This problem has now been addressed. After the fix, image captions are correctly positioned at the bottom of the image.
Fixed incorrect background color inheriting.

The background color defined for the body element could, in some cases, incorrectly impact the table header background color. This issue has been patched.

# Other changes

Improved the performance and memory consumption of the converter, making it even faster

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.17.0 (2024-06-19)

# Import from Word

# Features

Added support for captioned images

Captioned images exported by Export to Word 2.0 are now recognized in Import from Word and are properly converted to <figure> elements containing the caption in <figcaption>.

# Bug fixes

Fixed support for the “at least” line spacing option

When a paragraph in Word had line spacing set to some minimum value using the “at least” option, that spacing made the imported content unreadable.
CSS lacks a dedicated property for minimum line height; however, Import from Word converts such formatting to the max() CSS function, imitating the same feature from Word.

# Export to Word

Note:

The /v2/convert/html-docx endpoint has been introduced with this version.
The v2 provides a number of improvements, optimizations and fixes to the Export to Word feature. All changes mentioned in the changelog applies to the v2 version.
In case of any problem or regression we encourage you to use the previous version and report the discovered problems through the support channel.

# Features

Added Auto-detection to some built-in Word styles

Word has many predefined styles that can be explored using the Styles gallery. Export to Word can now automatically pick up some of them from the provided CSS style sheet and convert them to their respective style from Word. The following styles are currently supported:
- Normal
- Heading 1 through Heading 6
- Plain Text
- Quote
- Code
- Hyperlink
- Table Grid
Added table header rows support

Table rows inside the first <thead> are converted to header rows in generated Word documents. Header rows repeat at the top of each document page in Word.
Added table row height support

Table rows have their height preserved now, maintaining the original table layout.
Added support for table cell paddings

Table cell padding is supported and properly converted to table cell margins in this version of Export to Word.
Added support for more text-related CSS properties

Various CSS properties related to text convert more properly now. This includes properties such as font-weight, font-style, text-decoration, named font-size CSS properties (small, medium, large, etc.), and many more.
Added support for generic font families

CSS allows specifying generic font families, such as sans-serif or monospace, in the font-family property. Previously, they were treated as font names, but in this version, they are now properly recognized and transformed into some well-known font families, for example Times New Roman.
Added support of line spacing in paragraphs

The line-height CSS property is now converted to line spacing formatting in Word, ensuring that paragraph spacing better matches the original document.
Added support for spacing before and after paragraphs

The converter supports margin-top and margin-bottom CSS properties now and converts them to respective paragraph spacing.
Added support list-style-type defined in the CSS style sheet

The previous converter version supported the list-style-type property only if it was defined directly on the element via the style attribute. Now, list-style-type defined in the provided CSS style sheet is also recognized and supported.
Added support the font-family property in list items

font-family can be applied directly to <li> elements and will be preserved in generated Word documents.
Added support for break-before and break-after CSS properties

HTML with these properties will generate a page break accordingly, enabling more control over page layout.
Added support comments in more positions

In the release, comments attached to various elements are supported. This includes, among others, comments attached to tables, table cells, horizontal lines and block images.
Added time zone support in comment creation dates

Creation dates of comments take their time zone into account now, allowing Word to correctly display them in the user’s local time zone.
Added support for floating images

This version of Export to Word supports images with the float CSS property.
Added Base64 image support in headers and footers

Previously, when a header or footer contained an image with its source encoded in Base64, the converter did not preserve the image. Now, it correctly converts such images.
Added support for decorative images

Images without or with an empty alt attribute are now marked as decorative images.
Added support for image caption background color

The image caption now supports background color with an alpha channel.

# Bug fixes

Fixed paragraph background color not applying to the entire paragraph.

Before, the paragraph background color was applied only to the paragraph contents. Now, it’s added to the entire paragraph.
Fixed handling <span> elements with custom CSS classes

In the previous version, HTML documents containing <span> elements with custom class names didn’t convert properly. This issue has been resolved, and such documents are now generated correctly.
Fixed rendering of text with inherited font size

Previously, applying font-size: inherit to some text resulted in generating invisible text in the Word document. This was changed to display the text properly.
Fixed handling of invalid Unicode characters

Some Unicode characters are disallowed in Word documents and cannot be stored in them. The converter didn’t check their presence, which resulted in broken documents being generated. In this version, such characters are transformed into the � replacement character.
Fixed broken list editing in generated documents

Previously, when a generated document contained list items, their editing experience was broken when trying to indent them. That issue was resolved in this version.
Fixed support for empty table cells

When a table contained a cell without any content, or with empty elements (such as empty <span> elements), the generated documents were broken. In this converter version, such tables are converted properly.
Fixed conversion of alternative image text

In the previous version, images with the alt attribute were losing their alternative text during the conversion. That issue was addressed, and alternative text is now preserved.
Fixed hyperlinks with captioned images

Previously, when a hyperlink contained an image with a caption, the conversion resulted in a broken document. Now, the generated documents are valid.
Fixed support for suggestions in list items

In some cases, when a list item contained a text suggestion, it was possible for that suggestion to convert incorrectly, resulting in paragraphs and list items following the list item with the suggestion to be included in it. This release addresses this problem, and text suggestions in list items no longer include more content than they did in the source HTML.

# Other changes

Significantly improved conversion times

With this release, we notably sped up the Export to Word converter, making it more suitable for converting large HTML documents.
Extended support for image formats

This version of the converter adds support for the following image formats:
- image/apng
- image/avif
- image/bmp
- image/heic
- image/tiff
- image/webp
Improved image size support in all cases

Unless specified otherwise in the CSS style sheet, image size is no longer restricted to the document size. Additionally, when the converter cannot fetch an image, it will retain the size of the placeholder that is displayed in browsers instead of changing it to a fixed value.
Improved table background color conversion

The previous version of the converter did not always convert background colors in tables properly. In this version, detection of background color in tables has been changed to more closely match the HTML standard, so table cells in generated documents no longer lose their background colors in some cases.
Improved proper column width detection

This version of Export to Word correctly recognizes table columns created via <colgroup> and <col> and their styling, such as the width CSS property.
Improved table cell merging

Previously, merged table cells did not work correctly in all configurations. This has been improved to handle more complex table layouts correctly.
Improved conversion of percentages and relative units

Percentages and units such as em or rem are now more accurately calculated to more closely match how browsers render them.
Improved positioning of images

The converter more accurately converts image positioning now, including image captions.
Improved conversion of whitespace characters

Whitespace characters in HTML have complex rules determining whether such characters should be rendered. Previously, they were lost in some situations during the conversion. This release addresses this long-standing issue, and whitespace characters are now preserved correctly.
Improved support for nested comment threads and suggestions

When a deletion suggestion starts at the same place as a comment thread, the suggestion is now correctly preserved.
Improved the horizontal line support

Horizontal lines are now rendered as actual lines rather than very narrow boxes, ensuring they appear correctly in generated documents.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.16.3 (2024-04-30)

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.16.2 (2024-03-05)

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.16.1 (2024-01-09)

# Import from Word

# Other changes

Added the text-decoration-skip-ink CSS property to CSS resets
Added the font-size CSS property of HTML headings to CSS resets

# Export to Word

# Bug fixes

Fixed suggestions attribute breaking a conversion

The unsupported suggestions attributes are removed from the document.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

The above update patches the security vulnerability related to the heap buffer overflow risk.

# 1.16.0 (2023-11-14)

# Import from Word

# Features

Added support for newline suggestions

This release adds support for newline suggestions. This type of suggestion occurs in Word documents when a hard line break is inserted or deleted while changes are being tracked.

# Bug fixes

Fixed support for checkboxes in tables

Previously, when a table contained a checkbox inside a row or a cell, the converter didn’t recognize such a structure, and threw an error when performing some table operations. For example, when the cell containing the checkbox spanned several columns, the merging operation in the converter caused errors.
Fixed inheritance of complex formatting properties

Complex formatting properties, such as list numbering or underlines, weren’t always resolved correctly when they were defined on various levels of the style hierarchy. For example, paragraph styles that inherited a list definition from another paragraph style, but had a different list level specified, didn’t apply properly. This release adjusts resolving of such formatting.

# Export to Word

# Features

Added support for custom headers while fetching external images

Custom headers can now be added to every request which fetches external images.
It can be used e.g. to authorize access to external images.
The feature is available thanks to the extra_http_headers option.
You will find more information in our documentation.

# Bug fixes

Fixed incorrect rendering of images if fetching any of the images failed.

An image that couldn’t be loaded during the conversion is properly displayed as non-available in a Word document.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.15.1 (2023-10-05)

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

The above update patches the security vulnerability related to the heap buffer overflow risk.

# 1.15.0 (2023-09-19)

# Import from Word

# Features

Added CSS resets and extended formatting support

Browsers provide their own CSS stylesheets that apply default formatting to common HTML elements. Unfortunately, this formatting is incompatible with the default Word formatting. Previously, the IFW converter did not support overwriting conflicting browser styles. Now a new configuration option formatting.resets is available. It can be used to change selected formatting properties to the same values that are used in Word by default.

Along with this change, we introduced the new formatting configuration option that can be used to control the outputted formatting in a more granular way:
```
type ConversionConfig = {
    formatting?: {
        resets?: 'none' | 'inline';
        defaults?: 'none' | 'inline';
        styles?: 'none' | 'inline';
        comments?: 'none' | 'basic' | 'full';
    };
};
```
The old default_styles and collaboration_features.comments_styles options are now deprecated in favor of formatting.defaults and formatting.comments respectively.
Added default table styles

Tables in Word have some default formatting defined implicitly. For example, text in table cells is by default vertically aligned to the top, and tables themselves are aligned to the left. These properties differ from the default HTML formatting, where, for example, the content of table cells is by default centered. However, they are not saved anywhere in documents, so the converter added them as direct formatting to minimize the visual differences between Word and the HTML output. Unfortunately, this means that such default styles were always outputted regardless of whether the conversion of default formatting was actually enabled.

In this release, the implicit table formatting starts being recognized as the default formatting and is no longer added to the output if the formatting.defaults option is set to false. Along with this change, we also extended support for default table formatting, and table borders are now collapsed by default.

Please note that this may change the content produced by the converter when documents are converted with the conversion of default formatting disabled (which is the default case).

# Bug fixes

Fixed support for simple fields

Previously, some fields such as DATE could be incorrectly converted, resulting in content loss. Now, the converter will properly preserve the content of such fields.
Fixed escaping font names in CSS

When some part of a document was using a font with specific characters in its name, the converter outputted invalid CSS due to a lack of proper character escaping. This means that browsers ignored the outputted formatting and such fonts were not properly applied in the output. The Import from Word converter now properly escapes these characters.

# Export to Word

# Bug fixes

Fixed support for images inside span elements within tables
Fixed the rendering of documents with over 20,000 paragraphs
Fixed a problem with preserving multiple successive spaces in exported documents

Added a proper whitespace handling mechanism that preserves whitespace characters consistent with browser vendors.
Fixed incorrect handling of suggestions in lists when a suggestion was nested inside an inline element with no other content

The documents containing suggestions inside inline elements should no longer be broken.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.14.1 (2023-07-25)

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.14.0 (2023-07-11)

# Import from Word

# Features

Added support for the Strict Open XML format

The .docx files can be saved in the Strict Open XML mode, which differs significantly from the regular mode. This release adds support for DOCX files saved in the strict mode, resulting in better coverage of supported document formats.

# Bug fixes

Advanced Table Formatting: Last column formatting is not applied to merged table cells

Formatting of the last table column was ignored when the last table cell in a table row spanned multiple columns. Thanks to the fix, cells in the last column always have conditional formatting properly applied now.
Advanced Table Formatting: Formatting of corner cells in the last column is not applied to merged table cells

Formatting of corner cells in the last column was not applied when those cells spanned multiple columns. The fix should now ensure proper resolution of formatting for corner cells in the last column.
Advanced Table Formatting: Similar formatting regions should be mutually exclusive

When table cells belong to similar table regions, only formatting from one such region should be applied. For example, consider a table that contains a single cell. This cell belongs to both the first and the last column. However, in Word, only the formatting of the first column is applied to that cell. Import from Word follows the same behavior as Word now.
Table of Contents: Line breaks in headings are not properly supported

Headings in Word may contain soft line breaks, which are typically replaced by a single space character by default. However, the converter ignored those line breaks, resulting in a lack of whitespace in the output.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.13.0 (2023-05-30)

# Export to Word

# Features

Added support for resolved comments

Comments that are marked as comment.is_resolved will be resolved in a Word document.

# Bug fixes

Fixed generating Word documents with list suggestions

Suggestions in list items sometimes caused producing broken Word documents
Fixed broken conversion due to loading CSS

Some CSS syntax caused conversion to fail. The fix skips the invalid CSS properties.

# Import from Word

# Features

Added support for advanced table formatting

In Word, table styles can define the formatting of particular table sections, for example, the background color of cells in the last table row. In fact, many table styles built-in into Word use such formatting. Now, the Import from Word converter recognizes and properly converts it.
Added support for archived comments

Recently, CKEditor 5 released a feature called “comments archive” that allows archiving comments that are no longer necessary. Word, on the other hand, has an option that marks comments and entire comment threads as resolved. Until now, the Import from Word converter did not return any information on whether a comment is resolved or not, but since this release converted comments contain a new property called is_resolved that can be used to archive imported comments in CKEditor.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.12.0 (2023-04-28)

# Import from Word

# Features

Added support for table of contents

This release brings support for the table of contents. Currently, two possible formats of the table of contents are supported:
1. Automatic table of contents
2. Custom table of contents
Added support for horizontal alignment of non-inline images

In Word, there are two types of images: inline and floating. The latter can be freely positioned and have its other settings, such as text wrapping, modified. In this release, support for the horizontal alignment of floating images was added.

Note that HTML does not support a centered image with text wrapped around it, and thus importing such images does not preserve text wrapping.
Added support for 3D rotation via camera presets

Extending support for image rotation from the previous release: support for camera presets was added. Camera presets available from the 3-D Rotation tab of the Format Picture menu and containing rotation settings are converted properly now.

# Bug fixes

Fixed incorrectly aligned images

Disabling the config.default_styles option sometimes caused imported images to be misaligned. This release patched this issue and now images should have proper alignment regardless of the default_styles option.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.11.0 (2023-04-04)

# Export to Word

# Features

Added support for text indentation

The text-indent CSS property on paragraphs is properly recognized and applied to the Word document. Text indentation support is limited to absolute units, like pixels.

# Bug fixes

Fixed an issue where an invalid HTML table structure resulted in producing broken Word documents

Tables that are missing some part of the proper table structure, for example, empty table rows, are now properly handled by the converter.
Fixed an issue where block elements inside list items were producing invalid HTML

Block elements inside list items are converted to inline ones to preserve the correct list structure. Tables are extracted from lists as they are not supported in lists by Word.
Fixed an issue with fetching images from external servers

png and jpg images are now always downloaded and included in the `.docx`` file.

# Import from Word

# Features

Added support for conversion of SVGs

The Import from Word converter can correctly convert images using the SVG format. It supports both embedded and linked images.
Added advanced underline styling

Previously, the converter always outputted underline as a solid black line. Now, it supports underlines that use different colors or shapes. However, not all shapes available in Word are supported by CSS, hence some are transformed into the closest matching equivalents.

Advanced underline styling is represented using various text-decoration-* CSS properties. Additionally, please note that CKEditor 5 may need extra General HTML Support (GHS) plugin configuration to preserve it.
Added support for image rotation

Image rotation is now preserved by the converter. Both 2D and 3D rotations are supported. Currently, the default camera preset is supported.

Rotation is outputted as a transform CSS property, and thus it behaves slightly differently from Word: rotated images may overlap text in preceding and following paragraphs. This feature also needs an additional GHS configuration in order to work in CKEditor 5.
Added bookmarks support

Bookmarks are document regions associated with some name. They can be used, for example, to create links to fragments of a document. They are now supported by the Import from Word converter that represents them using an empty <a> element with the bookmark name saved in id and name attributes. Links to bookmarks are represented by URI fragments.

Note that, presently, CKEditor 5 does not support bookmarks.
Added support for custom unordered list bullets

Many Word documents contain custom bullets in unordered lists. The converter is used to transform them into the default circle bullet. Now, it preserves them in most common cases. However, non-standard symbol typefaces may sometimes produce unexpected results for now.

# Bug fixes

Trailing whitespace characters no longer preserve the background color

Some whitespace characters (such as the regular space or the em space) do not have their background color rendered by Word when they are at the end of a text line. The converter now detects such characters when they are located either at the end of a paragraph or before a soft line break and strips background styling from them. Note that the converter cannot detect trailing whitespace characters at the end of arbitrary text lines as of now.
Emoji sequences with attached comments are no longer broken into parts

Previously, the converter split emoji sequences into individual parts when a comment was attached to them. For example, the 👩‍👩‍👦‍👦 emoji sequence with a comment attached to it would be split into 👩👩‍👦‍👦. It is correctly preserved now.
Fixed invalid conversion of list definitions using none numbering format

List items that use the none numbering format and that do not have any additional numbering text are no longer outputted with the default circle bullet, but rather have no marker at all.
Fixed parsing of external images

Images that were not embedded into a document itself used to cause conversion errors. Now, the converter correctly supports them.
Fixed comments conversion in empty tables

Until now, comments attached to empty tables were not preserved. This is no longer the case. However, note that there’s still an open issue regarding comments in empty tables in CKEditor 5.
Preserved level definitions in multi-level lists

Skipped list levels are no longer represented by lists with invisible markers, but rather by particular list items with invisible markers. This improves the editing experience of such lists in CKEditor because items inserted at previously skipped levels have their marker formats preserved now.
Improved handling of spaces

The Import from Word converter follows the space behavior in HTML more strictly now. It also no longer eagerly transforms regular spaces into non-breaking ones. This improves line-wrapping behavior because in some cases, the styled text could cause unnecessary line wraps.
Improved HTML output for collaboration features

The HTML produced for collaboration features has been optimized to reduce unnecessary styling duplications over content that has the same visual effect.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.10.0 (2023-02-07)

# Export to Word

# Features

Added better support for complex suggestions nesting

Significantly improved support for nested suggestions of both the insertion and deletion types.

# Bug fixes

Fixed nested suggestions generating a broken document

Nested suggestions of insertion and deletion types sometimes resulted in producing broken documents. After the fix, complex suggestions nesting is more stable.
Fixed various conversion issues caused by nesting content inside div elements

In some cases, nesting content inside div elements resulted in producing broken documents or applying incorrect style over nested inline content. After the changes, content nested in div elements is properly recognized.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.9.0 (2022-11-29)

# Export to Word

# Features

Added the base_url conversion option

This option allows setting the base URL for resolving images’ and links’ relative URLs in the content, e.g. setting the base_url option to https://ckeditor.com will result in resolving the /docs link as the https://ckeditor.com/docs one.
Added support for SVG images

SVG images are now supported by the converter, both as base64 and URL links.

# Bug fixes

Fixed the issue where conversion of the style element inside other elements breaks conversion.

After the fix, the style element can be placed inside other elements, e.g. a div and the CSS styles will be properly preserved.

# Import from Word

# Features

Added the config.collaboration_features.comments_styles configuration option

It allows to choose one of the existing presets (none, basic, full), allowing to configure which comments’ styles should be converted.

# Bug fixes

Fixed the issue where in rare cases imported tables had incorrect width
Fixed the issue where in rare cases text background turned black
Fixed incorrect parsing of suggestions that is applied to a link spanning over multiple blocks
Fixed emoji groups being split during the conversion

# Other changes

Reduced HTML payload size and improved its quality

It was done by merging side CSS properties to CSS property shorthands.
As an example, properties such as border-top, border-right, border-bottom, and border-left will be reduced to a single border property.

# Common changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.8.1 (2022-11-08)

# Bug fixes

Fixed incorrect log messages while using an unlimited license

While using an unlimited license, notifications with information about overusage were present in the application logs.
Now the logs are present only after exceeding the limits in case of a limited license.

# 1.8.0 (2022-11-02)

# Features

Added support for Import from Word

Import from Word allows to convert a Word file into HTML code. Comments and suggestions will also be extracted.
You will find more information about the feature in our on-line documentation.
The feature requires a separate license. To start using it, please contact our sales or subscribe to our free 30-day trial.

# Other changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.7.4 (2022-10-18)

# Other changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.7.3 (2022-09-06)

# Bug fixes

Fixed incorrect handling of page breaks inside table cells

Page breaks inside table cells were breaking document conversion. After the fix, page breaks are properly recognized in the produced Word documents.
Fixed conversion fails if figcaption was placed under a figure with a class attribute that lacks image or table class name

When figcaption was placed under a figure with a class attribute that lacks an image class name, the conversion failed. After the fix, the conversion is successful and the figcaption is placed in the Word document.
Fixed conversion fails if a figcaption was placed inside a table figure element

After the fix, table figcaption elements are ignored, no longer causing the conversion to fail.

# Other changes

Improved error messages

Messages returned by document conversion provide better debugging experience to integrators if the conversion failed.
Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.7.2 (2022-07-26)

# Other changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.7.1 (2022-06-14)

# Other changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.7.0 (2022-05-05)

# Features

Added support for specifying width of table cells and columns

The width of table cells can be specified by the width style and the colgroup element.
Added support for ARM64v8 CPU architecture

The Export to Word On-Premises Server can now be run on machines with the ARM64v8 CPU architecture.

# Other changes

Updated dependencies

Selected dependencies have been updated to the latest versions, improving performance and security.

# 1.6.0 (2022-03-22)

# Features

Introduced auto paragraphing for orphan nodes

Orphan text or inline nodes are automatically wrapped with paragraphs.
Added support for comments inside table cells

Table cells show commented cell content if added.
Added support for suggestions inside table cells

Table cells show cell content suggestions if added.
Improved features compatibility for older (<365) MS Word versions

Documents with images in base64 format and tables are now correctly supported when opening documents with older MS Word versions (<365).

# Bug fixes

Fixed empty href in links breaking conversion
Fixed broken documents caused by invalid markup for comments
Fixed an issue where code snippets in table cells have an additional paragraph under a snippet
Fixed an issue where blockquote elements including plain text content broke document conversion

# Other changes

Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.
Cleaned up dependencies
Improved error messages

Some of the errors now contain more detailed messages, including explanations and possible actions to solve the problem.

# 1.5.2 (2022-02-08)

# Other changes

Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.

# 1.5.1 (2022-01-18)

# Other changes

Update Node.js to version 16
Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.

# 1.5.0 (2021-11-16)

# Features

Added support for autoparagraphing plain text comments

Comment content property was only accepting HTML content. From now on, it is possible to use plain text also, which helps with custom CKEditor 5 comments integration.

# Bug fixes

Fixed incorrect generation of suggestions for new paragraphs

# Other changes

Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.
From this version on, the Docker Engine 20.10 or higher is recommended.
Updated base docker image (Alpine Linux) from version 3.12 to 3.14
Added an orientation parameter to the demo page

# 1.4.0 (2021-10-05)

# Features

Added support for custom numbering in lists

List styles supported in CKEditor 5 are also supported by the converter.
Enabled starting ordered lists from number other than 1

Starting an order list from number other than 1 can be accomplished by providing a [start] attribute.
Added support for list indentation

You can control the indentation of list elements by providing margin style.
Added support for table and table figure width

You can control the width of table elements by providing width attribute or style.
Added support for page orientation

Page orientation can be set using the options.orientation property, accepting portrait (default) and landscape values.

# Bug fixes

Fixed documents broken due to inserting empty table cells or table cells with multiple spaces
Fixed conversion of images that are not inside a paragraph or a figure element
Fixed issue with suggestions spanning two paragraphs instead of ending on the first one
Fixed incorrect generation of suggestions for new paragraphs

# Other changes

Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.

# 1.3.0 (2021-08-24)

# Features

Increased maximum size of the HTML

The service allows to convert HTML with the maximum size of the content 20 MB.

# Other changes

Improved logs readability
Updated dependencies

Updated selected dependencies to the latest versions - improved the performance and security.

# 1.2.1 (2021-04-04)

# Features

Introduced support for inline images

Inline images (images that are inside a paragraph) are now correctly rendered.
Introduced support for nested tables

The converter now supports HTML with nested tables (tables inserted into other tables’ cells).

# Bug fixes

Fixed the header/footer CSS validation. CSS is now not obligatory
Fixed adjacent hrs rendering
Fixed the generation of documents containing div elements with table elements inside
Fixed the API section in the documentation

# Other changes

Cleaned up logs

Removed stack traces, fixed data serialization and improved messages.

# 1.1.0 (2021-02-25)

# Features

Added validation for document’s margins

Supported units: in, cm, mm, px.
Added support for auto pagination feature

The auto pagination feature is enabled by default now.
Added a creation date for comments

The created param is required for comments now.
Added the timezone option for a document

Use the timezone option to set a document timezone. For example, if you use Comments and suggestions feature, created date time metadata will include the provided timezone option or UTC if the timezone option has not been set.
Added support for tables alignment

Enabled support to align tables to left and right using CSS float property.
Added support for comments in suggestion threads

Comments inside suggestion threads are preserved as a separate comment thread on the same range as suggestion.
Added support for the page number and total pages in document headers and footers

Headers and footers support dynamic data like current page number and total page number. It can be inserted by using the special placeholder element in header/footer HTML. For current page number use and for total page number.

# Bug fixes

Fixed block quotes nested inside a table breaks a document
Fixed Horizontal line after or before a code block is rendered inside it
Fixed the styled text inside a link causes a link loss

# Other changes

Improved various document formatting to make it more similar to CKEditor 5 content
Improved demo page

The demo page is now more readable and allows to use all available converter’s features.
Updated documentation

# 1.0.0 (2020-08-25)

# Features

Added On-Premise version of DOCX Converter.

The service allows to convert HTML into DOCX documents via the REST API.
Added authorization.

JWT authorization can be set with SECRET_KEY environment variable.