1. User guide
  2. Matecat utilities

Matecat release notes

13/12/2023

New features

  • Added APIs for the creation of custom billing models that can be used in projects: for each analysis bucket, you can now decide to use different payable rates than Matecat's default ones (e.g. discounting the words in 100% matches to 20%, rather than 30%).
  • Added the possibility to create projects with character limits or notes for segments.
  • Added shortcuts for curly quote signs and apostrophes.
  • Added support for Firefox.
Updates
  • Heavily reworked editor interactions in order to fix bugs and make improvements especially when it comes to tag management.
  • Updated Matecat's management of XLIFF 2.0 files in order for states in target XLIFF files to be consistent with the states in Matecat's editor.
  • Added clear error messages when trying to import invalid TMX files.
  • Added the Mam, Kanjobal and Chuj languages.
  • Added indicator of number of resolved comments in a job.
  • Improved glossary upload error messages.
  • Updated Matecat's character counter in order to include Sinhala characters among those counted based on their UTF-16 byte size, making it fully compliant with Google Ads.
  • Updated glossary export process to provide an xlsx file rather than a csv as the output.
  • Updated glossary search to make it more flexible and user-friendly.
  • Updated password reset flow to make it more user-friendly.

Fixes
  • Fixed a bug related to the deletion of TM matches from memory keys by users different than the key's owner.
  • Fixed a bug related to segments with no translatable content not being detected as 101% matches.
  • Fixed a bug related to intra-document fuzzy matches not being placed in the correct analysis bucket with MT active.
  • Fixed a bug related to whitespace mismatches around tags not being flagged correctly.
  • Fixed a bug related to the upload of files directly from Google Drive.
  • Fixed a bug related to the underlinement of composite glossary terms.
  • Fixed a bug related to the guess tags function deleting formatting signs at the end of segments.
  • Fixed a bug related to segmentation rules not being applied correctly upon project creation.
  • Fixed a bug related to the AltLang integration not functioning correctly.

30/08/2023

New features
  • Implemented ModernMT's revolutionary context-awareness technology for all users with an active ModernMT subscription. More details here.
  • Added an AI assistant button to request a GPT-powered explanation for the highlighted source word(s). More details here.
  • Added a button to quickly insert the highlighted source word(s) in the glossary.
  • Added support for segments states in xliff 2.0 files.
  • Added a clear debug message for the situations where MT and TM matches cannot be retrieved due to external factors such as firewalls or antivirus software.
Updates
  • Updated all dates in APIs to ISO format.
  • Updated project creation API to require API key authentication.
  • Added 68 new languages in order to support all of ModernMT's 200 languages.
  • Updated analysis algorithm to count variables as one word in order to make word count more precise.
  • Added support for variables formatted as ${placeholder}.
  • Added support for Lokalise universal placeholders.
  • Expanded sprintf variable support.
Fixes
  • Fixed a bug that prevented drafts from being downloaded in the presence of tag issues in unconfirmed TM matches.
  • Fixed a bug that caused glossary terms in the source not to be matched due to the presence of more than one punctuation character right after them.
  • Fixed a bug that prevented the addition of domain and subdomain to a glossary term if the glossary file did not already have at least a domain in it.
  • Fixed a bug that prevented users from completing their order when outsourcing jobs to Translated.
  • Fixed a bug that caused visualization issues in the case of long names for glossary keys.
  • Fixed a bug that caused html tags with colons in them not to be tokenized.
  • Fixed a bug that caused some characters to be shown differently in the source and target text portions of the "TM Search" tab.
  • Fixed a bug that caused notes from xliff 1.2 files not to be shown in some particular cases.
  • Fixed a bug related to double-encoded greater-than and less-than signs.

30/03/2023

Updates

  • An all-new version of the fast analysis (Matecat's component that counts the number of words in source files and looks for repetitions and internal matches) has been released, quicker and more accurate than ever.
  • The behavior of the glossary tab has been changed in order for the terms to appear in the order in which they're found in the source segment.

  • A component has been added to Matecat in order to track most of the errors happening during use, which has already allowed us to correct a few unnoticed bugs and that will allow us to catch any issues faster than ever.

  • The interaction for the selection of target languages has been improved, making it more intuitive and user-friendly.

  • An improvement has been made to the glossary upload process: the glossary now ignores any whitespaces found at the beginning or end of terms.

Fixes
  • Fixed a bug related to the addition of terms to glossaries when the target language is Canadian English.

  • Fixed a bug that prevented the upload of xliff files extracted from Crowdin and other tools to Matecat.

  • Fixed a bug that made glossaries added to jobs by people different than the job's owner visible to anyone opening the link.

  • Fixed a bug that prevented the deletion of TM matches from functioning correctly.

  • Fixed a bug that was causing incorrect tag issues to be flagged in Airbnb smart count segments.

  • Fixed a bug that prevented the copy-source-to-target function from working in split jobs.

  • Fixed a bug to restore OMP's function for opening Matecat's homepage with some settings already configured.

23/02/2022

New features

  • We have released our all-new glossary, which is now more powerful than ever. For any doubts you might have about it, you can refer to our user guide
  • We have released the possibility for our enterprise customers to send us character limits for strings and have the length of the target translation checked against the limit with automatic QA

  • We have added the possibility to delete comments left on segments in addition to the possibility of marking the whole thread as solved

Updates
  • We have expanded the integration between Matecat and ModernMT for our users to have more options than ever when connecting a license

  • The character counter now counts CJK characters and emojis based on their byte size, according to industry standards

  • The UX for target-language choice has been updated to make it quicker for everyone, let us know if you like it!

  • We have updated some of the copies in the editor and "MyProjects" page in order to make them clearer

  • We have updated the UI of the "Translation Memory and Glossary" tab of the "Settings panel" to provide better UX when creating new memory keys or activating shared ones

  • We have expanded placeholder support to the placeholders used for Apple code, which is becoming ever more popular among our customers, and to other placeholders requested by our customers

  • We have increased the upload limit for tmx files from 100 to 300 mb

  • We have added new languages to LexiQA

    • Armenian (hy-AM)
    • Azerbaijani (az-AZ)
    • Belarusian (be-BY)
    • Haitian Creole (ht-HT)
    • Kazakh (kk-KZ)
    • Kirundi (rn-BI)
    • Kurdish Kurmanji (ku-TR)
    • Kurdish Sorani (cb-IQ)
    • Kyrgyz (ky-KG)
    • Nepali (ne-NP)
Fixes
  • Fixed a bug that caused Matecat to return the error "Invalid upload token" after every upload error and was causing frustration and slow-downs for our users

  • Fixed a bug regarding tag mismatch issues being shown as tag order mismatch issues

  • Fixed a bug regarding the capitalization button not working for non-Latin scripts

04/10/2022

New features

  • Added possibility to upload multiple tmx/glossary files from "Settings" tab
  • Added setting to activate counter for target segment characters
  • Added possibility to search inside own chunk/whole job for split jobs
  • Added possibility to search for non-breaking spaces
  • Added possibility to delete jobs and projects permanently from the "My Projects" page
  • Added support for word joiner character + related keyboard shortcut
  • Added ID number to eponymous ph tags in order to reduce occurrence of tag issues
  • Added support for jsont2 format for the new TOS API
    Updates
    • Added new languages to LexiQA:

      • Latin-American Spanish (es-419)
      • United States Spanish (es-US)
      • Assamese (as-IN)
      • Georgian (ka-GE)
      • Javanese (jv-ID)
      • Sinhala (si-LK)
      • Uzbek (uz-UZ)
    • Worked with LexiQA to reduce occurrence of false-positives in the presence of tags
    • Improved UX for selection of multiple target languages

    • Set "Messages" editor tab to open automatically when present

    • Changed state of pre-translated segments from bilingual files from "translated" to "approved" to better cover most-frequent use case

    • Replaced "Outsource" button with "Contact us" button for languages not supported by Translated

    • Improved clarity of confirmation pop-up shown when making changes to an autopropagated segment

    • Improved UX when performing bulk translate/approve actions

    • Added label to TM matches indicating language variant

    Fixes
    • Fixed a bug that made the "Share resource" button for MyMemory keys show an error message even though the operation had been succesful

    • Fixed a bug that made the analysis page for a job inaccessible after changing passwords for that job

    • Fixed a bug that prevented the shortcut panel from opening

    • Fixed a bug that prevented users from scrolling down the list of members for a team

    • Fixed a bug related to the integrity of xliff 1.2 files downloaded from Matecat

    • Fixed a bug that prevented users from opening R2 links from "My Projects" page right after creating them

    • Fixed a bug that prevented users from scrolling down to the end of a link when a comment was present on the first segment of the job

    • Fixed a bug that caused Matecat to show a 403 page when opening a job in archived/cancelled states

    • Fixed a bug that caused reviewed words to be calculated wrong in the case of massive use of the "bulk approve" function

    • Fixed a bug that prevented users from seeing the whole text of closed segments when using Safari

    • Fixed a bug that caused the split function to behave oddly when selecting text

    • Restored the correct error message when uploading doc files with revisions/comments

    • Fixed a bug that prevented the "download translation" button from appearing in jobs with only ICE matches

    • Fixed a bug that prevented LexiQA from being activated in multilingual projects with one language not supported

    16/02/2022

    New features
    • Button added to edit interface to show the first segment in the active chunk

    Updates
    • Increased the number of language combinations supported by the "guess tags" function to 51

    • Changed behaviour when clicking progress bar or "to-do" section in completed jobs: until now, clicking the progress bar or the "to-do" section in a 100% complete job would redirect to the first segment of the job. The behaviour has now been changed to show a message about the absence of unconfirmed segments in the job.

    Fixes
    • Fixed a bug that prevented Matecat from replicating line breaks in target Gdoc files in Simplified Chinese and Japanese

    • Fixed a bug that caused the page to go blank when opening the "messages" tab for a segment on Safari

    • Fixed a bug in the "search and replace" function when using the " symbol

    • Fixed two bugs regarding the cross-language matches function

    • Fixed a bug that prevented autopropagation from working correctly for ICE matches

    • Fixed a bug that made an unwarranted "no connection" alert appear sometimes upon propagating segments

    23/12/2021

    Fixes
    • Improved TM search visualization: now the language of the match is displayed in the source details.

    • Modified the message of the feedback submission popup.

    • Fixed a bug related to the comments feature in the revision issues box.

    • Fixed a bug related to private TMs settings

    • Fixed a bug related to the split segments feature.

    • Fixed a bug related to the project name in non-Latin characters.

    02/12/2021

    New features
    • Kurdish Sorani and Serbian Latin MT enabled

    • Added new subjects: adwords and European governmental policies

    • Implemented export of the QR in .csv: available for jobs < 90K words

    • Edge is now officially supported.

    Fixes
    • Fixed a bug related to HTML tags.

    • Fixed a bug related to < and > symbols in tags.

    • Fixed a bug related to the glossary feature: now it is possible to delete and edit terms from the editor page.

    • Fixed a bug related to the outsource page.

    • Fixed a bug related to the editing of ICE match in revise mode

    18/10/2021

    Fixes
    • Improved management and download of files with emojis.

    • Improved management of tags with negative IDs.

    • Fixed a bug related to the error messages visualization: users were redirected to a 404 error page and the correct error message was not displayed.

    • Updated error message when Matecat is unable to create the project.

    • Improved filters for management of XML files.

    • Improvement management and upload of protected PDF files.

    • Improved management of href tags.

    • Fixed an issue related to the visualization of revised segments in the QR.

    07/09/2021

    Fixes

    • Improved management and download of files with emojis.

    • Fixed a bug related to tab signs: now they are correctly displayed in the source segments.

    • Fixed a download issue that occurred when there is a source file with no segments to translate.

    • Improved management of XLIFF files with too long trans-unit IDs.

    • Improved management and download of XLIFF 2.0 files.

    • Fixed some minor bugs related to the comments feature in the editor page.

    • Fixed an issue related to the revision of draft segments during revise mode.

    11/08/2021

    Fixes
    • Fixed a bug related to LexiQA not showing any error;

    • Fixed a bug related to angle brakes: now their presence in the source file don’t cause any tag error;

    • Modified behavior for Korean language: now MateCat adds automatically a trailing space at the end of a sentence;

    • Fixed a bug related to the target file name when the file name was < three characters;

    • Subfiltering: for the Hungarian language only, now “-ot” suffix is not blocked as tag when next to the symbol %;

    • Fixed a bug related to the analysis of jobs without MT enabled;

    • Fixed an issue with missing spaces in SDLXLIFF target files;

    • Fixed a graphical glitch related to revise mode: now the popup warning that there are still some segments in draft is correctly displayed.

    14/07/2021

    Fixes
    • Subfiltering: fixed a bug related to some tags not blocked in the UI.

    • Fixed an issue with the “replace all” feature in chunks: now the “replace all” feature does not apply to the entire job but only to the single chunk.

    • Fixed a bug related to the split job feature: now users can split smoothly a job from the “my projects” panel.

    • Modified the “preview” label in the “my projects” panel: it has been changed into “draft”.

    • Fixed a glitch that caused the opening of two tabs when clicking on the “Open QR” button in the editing page.

    • Fixed a bug related to the glossary: now it is possible to add terms while working on a project.

    • Enabled LexiQA for ta-IN locale.

    • Quality report: improved the visualization of issues in R2.

    • Fixed a bug related to the display of the volume analysis of multilingual projects.

    • Fix for a bug related to the comment feature in split jobs: now translators can leave comments chunks not assigned to them.

    • Fix for a graphical bug related to the volume analysis of split jobs.

    24/06/2021

    Fixes

    • Fix for “Export XLIFF” feature with .zip files with missing target field: now XLIFF files are exported correctly from Matecat.
    • Fixed for a bug related to < sign inside a placeholder: now it is managed properly by MateCat.
    • Elements inside single curly brackets are not blocked anymore.
    • Fix a bug related to the replace all feature when replacing " /" with "/".
    • Fixed an issue related to the option “whole word” of the replace feature.
    • Fix for extra half space at the end of the segment in Japanese and Chinese.

      01/06/2021

      Updates
      • Words limit per project set to 250k words.

      • Files limit per project set to 100 files.

      Fixes
      • Fix for PPTX and PPT files download: now newlines in segments don't break the target file and no error messages are displayed.

      • Fix for bugs related to % sign wrongly blocked as tag.

      • Fix for a bug related to quotation marks in PO files.

      • Fix for a bug related to some keyboard shortcuts.

      19/05/2021

      New features
      • You can now change the job password of revise stage.

      Updates
      • Improved management and download of files with emoji

      Fixes
      • Fixed an issue that returned “Not Found” when clicking the “details” button next to ModernMT from the machine translation tab in the settings

      • Fixed a bug related to the glossary: now it works correctly when modifying an entry.

      • Fixed a bug related to the replace all feature: now after the page refresh, users are not moved to a random segment.

      • Fix for Guess Tag: if I click on “copy source to target”, tags are correctly placed.

      • Fix for the API link in the homepage footer: now it redirects to the correct page.